Monitoring method and device for service system
Through machine learning models, the monitoring data of the power business system is predicted and the threshold range is dynamically adjusted to identify abnormal data, which solves the problems of false alarms and failure to detect abnormalities in traditional monitoring methods, and improves system stability and monitoring performance.
Patent Information
- Application Number
- CN202411105392.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-13
- Publication Date
- 2024-12-03
AI Technical Summary
Traditional power business system monitoring methods use fixed thresholds to detect the operating status of the system, resulting in false positives easily when the system operation parameters change, and real abnormalities cannot be discovered in time.
The machine learning model is used to predict the monitoring data of the power business system, dynamically adjust the threshold range, and identify abnormal data by comparing the monitoring data with the dynamic threshold range.
It improves the accuracy of monitoring data, avoids false alarms, enhances the stability and security of the system, adapts to changes in operating status, and improves the overall monitoring performance.
Smart Images

Figure CN119065923A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of electronic digital data processing, and in particular to a monitoring method and device for a business system. Background Art
[0002] Traditional power business system monitoring methods usually use fixed monitoring thresholds to detect the operating status of the system. These thresholds are pre-set based on historical data and experience and remain unchanged during the monitoring process.
[0003] However, in actual applications, the operating status of the system will be affected by many factors, which will cause changes in the operating parameters of the system. In this case, the use of fixed thresholds is prone to false alarms, which may cause normal data to be misjudged as abnormal, or the real anomaly may not be discovered in time. Summary of the invention
[0004] The technical problem to be solved by this application is to provide a monitoring method and device for a business system, which can accurately identify whether the monitoring data of the power business system is abnormal. The specific solution is as follows:
[0005] A monitoring method for a business system, comprising:
[0006] Acquire monitoring data of the power business system within a preset time, as well as historical monitoring data of a time period before the preset time;
[0007] Inputting historical monitoring data of a time period before the preset time into a pre-trained machine learning model to obtain a prediction result output by the machine learning model, wherein the prediction result includes a predicted value of the monitoring data of the power business system at the preset time;
[0008] Determine the dynamic monitoring threshold range corresponding to the preset time according to the monitoring data prediction value;
[0009] Comparing the monitoring data with the dynamic monitoring threshold range;
[0010] When the monitoring data is not within the dynamic monitoring threshold range, it is determined that the monitoring data is abnormal.
[0011] In the above method, optionally, the training process of the machine learning model includes:
[0012] Acquire an initial machine learning model to be trained and a training data set; the training data set includes multiple training samples and a sample label of each training sample; each training sample includes monitoring data of a historical time period, and the sample label of each training sample includes monitoring data of the next time of the historical time period;
[0013] Using the training samples in the training data set to update the model parameters of the initial machine learning model to train the initial machine learning model;
[0014] When the initial machine learning model meets the preset training completion conditions, the initial machine learning model that meets the training completion conditions will be used as the trained machine learning model.
[0015] In the above method, optionally, determining the dynamic monitoring threshold range corresponding to the preset time according to the monitoring data prediction value includes:
[0016] Obtaining error parameters corresponding to the monitoring data;
[0017] The dynamic monitoring threshold range corresponding to the preset time is determined according to the monitoring data prediction value and the error parameter.
[0018] The above method, optionally, after determining that the monitoring data is abnormal, further includes:
[0019] Determining an abnormal level of the monitoring data;
[0020] Determining a display position of the monitoring data in a preset alarm page according to the abnormality level of the monitoring data;
[0021] The monitoring data is displayed at the display position in the alarm page.
[0022] The above method may, optionally, after displaying the monitoring data at the display position in the alarm page, further include:
[0023] In case of receiving a processing operation for the monitoring data, determining a classification result of the monitoring data according to the processing operation, the classification result representing whether the abnormality determination is correct or incorrect;
[0024] The monitoring data is stored according to the classification result of the monitoring data.
[0025] A monitoring device for a business system, comprising:
[0026] An acquisition unit, used to acquire monitoring data of the power business system within a preset time, and historical monitoring data of a time period before the preset time;
[0027] A prediction unit, used to input the historical monitoring data of the time period before the preset time into a pre-trained machine learning model to obtain a prediction result output by the machine learning model, wherein the prediction result includes a predicted value of the monitoring data of the power business system at the preset time;
[0028] A first determining unit, configured to determine a dynamic monitoring threshold range corresponding to the preset time according to the monitoring data prediction value;
[0029] A comparison unit, used for comparing the monitoring data with the dynamic monitoring threshold range;
[0030] The second determining unit is configured to determine that the monitoring data is abnormal if the monitoring data is not within the dynamic monitoring threshold range.
[0031] In the above device, optionally, the prediction unit includes:
[0032] An acquisition subunit is used to acquire an initial machine learning model to be trained and a training data set; the training data set includes multiple training samples and a sample label of each training sample; each of the training samples includes monitoring data of a historical time period, and the sample label of each training sample includes monitoring data of the next time of the historical time period;
[0033] A training subunit, used to update the model parameters of the initial machine learning model using the training samples in the training data set to train the initial machine learning model;
[0034] The execution subunit is used to use the initial machine learning model that meets the preset training completion conditions as the trained machine learning model when the initial machine learning model meets the preset training completion conditions.
[0035] In the above device, optionally, the first determining unit includes:
[0036] An acquisition subunit, used to acquire error parameters corresponding to the monitoring data;
[0037] The first determination subunit is used to determine the dynamic monitoring threshold range corresponding to the preset time according to the monitoring data prediction value and the error parameter.
[0038] The above device may optionally further include:
[0039] A third determining unit, configured to determine an abnormality level of the monitoring data;
[0040] a fourth determining unit, configured to determine a display position of the monitoring data in a preset alarm page according to an abnormality level of the monitoring data;
[0041] A display unit is used to display the monitoring data at the display position in the alarm page.
[0042] The above device may optionally further include:
[0043] A receiving unit, configured to, upon receiving a processing operation for the monitoring data, determine a classification result of the monitoring data according to the processing operation, wherein the classification result represents whether the abnormality determination is correct or incorrect;
[0044] A storage unit is used to store the monitoring data according to the classification result of the monitoring data.
[0045] The present application provides a monitoring method and device for a business system, the method comprising: obtaining monitoring data of an electric power business system within a preset time, and historical monitoring data of a time period before the preset time; inputting the historical monitoring data of a time period before the preset time into a pre-trained machine learning model to obtain a prediction result output by the machine learning model, the prediction result comprising a predicted value of the monitoring data of the electric power business system at the preset time; determining a dynamic monitoring threshold range corresponding to the preset time according to the predicted value of the monitoring data; comparing the monitoring data with the dynamic monitoring threshold range; and determining that the monitoring data is abnormal if the monitoring data is not within the dynamic monitoring threshold range. Through the above scheme, abnormal monitoring data can be identified more accurately, effectively avoiding false alarms, thereby improving the stability and security of the system. By dynamically adjusting the threshold range, it is possible to better adapt to the electric power business system with a constantly changing operating state, thereby improving the overall monitoring performance. BRIEF DESCRIPTION OF THE DRAWINGS
[0046] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.
[0047] Figure 1 A method flow chart of a monitoring method for a business system provided in this application;
[0048] Figure 2 A flowchart of the training process of a machine learning model provided in this application;
[0049] Figure 3 A flowchart of a process for determining a dynamic monitoring threshold range corresponding to a preset time provided by the present application;
[0050] Figure 4 An example diagram of an alarm page provided for this application;
[0051] Figure 5 A schematic diagram of the structure of a monitoring device for a business system provided in this application. DETAILED DESCRIPTION
[0052] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.
[0053] In this application, the terms "comprises", "comprising" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprising a ..." does not exclude the presence of other identical elements in the process, method, article or device comprising the element.
[0054] The embodiment of the present application provides a monitoring method for a business system, which can be applied to an electronic device, which may include a personal computer, a server, a tablet device, a mobile phone, or a smart wearable device, etc. The method flow chart of the method is as follows: Figure 1 As shown, specifically including:
[0055] S101: Acquire monitoring data of a power business system within a preset time, and historical monitoring data of a time period before the preset time.
[0056] In this embodiment, the monitoring data may include system operating parameters. The system operating parameters may include one or more of response time, throughput, number of concurrent users, system available time, failure rate, recovery time, access control indicators, resource utilization, interaction fluency, page loading time, etc. Resource utilization may include one or more of CPU utilization, network bandwidth utilization, and memory utilization. The recovery time is the time required for the system to recover from a faulty state to a normal state.
[0057] In some embodiments, the monitoring data may also include power business parameters, which may include power consumption, peak and valley values of power consumption, and power load.
[0058] Optionally, the time period before the preset time may be a time period such as one year, six months, one month, fifteen days, or seven days before the preset time.
[0059] S102: Input the historical monitoring data of the time period before the preset time into a pre-trained machine learning model to obtain a prediction result output by the machine learning model, wherein the prediction result includes the predicted value of the monitoring data of the power business system at the preset time.
[0060] In this embodiment, the machine learning model corresponds to the monitoring data, and different monitoring data corresponds to different machine learning models.
[0061] Optionally, the machine learning model can be a CNN-LSTM model, that is, the machine learning model can be composed of a convolutional neural network CNN, a time recurrent neural network LSTM and a fully connected layer, wherein CNN is used to extract spatial features in historical monitoring data, and LSTM can be used to extract temporal features in historical monitoring data; thereby, the fully connected layer can combine the spatial features and temporal features in the historical monitoring data to obtain accurate monitoring data prediction values.
[0062] Optionally, the historical monitoring data of the previous time period is data in which no abnormality occurs in the previous time period.
[0063] S103: Determine a dynamic monitoring threshold range corresponding to the preset time according to the monitoring data prediction value.
[0064] In this embodiment, the dynamic monitoring threshold range may include a dynamic monitoring upper threshold and a dynamic monitoring lower threshold.
[0065] S104: Compare the monitoring data with the dynamic monitoring threshold range.
[0066] Optionally, the monitoring data may be compared with a monitoring upper threshold and a dynamic monitoring lower threshold in a dynamic monitoring threshold range, respectively, to determine whether the monitoring data is within the dynamic monitoring threshold range.
[0067] S105: When the monitoring data is not within the dynamic monitoring threshold range, determine that the monitoring data is abnormal.
[0068] In this embodiment, if the monitoring data is not within the dynamic monitoring threshold range, it indicates that the difference between the monitoring data and the predicted value of the monitoring data is large. In this case, it can be determined that the monitoring data is abnormal.
[0069] By applying the method provided in the embodiment of the present application, abnormal monitoring data can be more accurately identified, thereby improving the stability and security of the system. By dynamically adjusting the threshold range, the power business system with constantly changing operating conditions can be better adapted, thereby improving the overall monitoring performance.
[0070] In one embodiment provided in the present application, based on the above solution, optionally, the training process of the machine learning model is as follows: Figure 2 As shown, specifically including:
[0071] S201: Obtain an initial machine learning model to be trained and a training data set; the training data set includes multiple training samples and a sample label for each training sample; each of the training samples includes monitoring data for a historical time period, and the sample label for each of the training samples includes monitoring data for the next time of the historical time period.
[0072] S202: Using the training samples in the training data set to update the model parameters of the initial machine learning model to train the initial machine learning model.
[0073] In this embodiment, a target training sample currently used for training can be selected from each training sample in the training data set; the monitoring data of the historical time period included in the target training sample is input into the initial machine learning model to obtain a predicted value of the monitoring data for the next time in the historical time period; the loss function value is calculated by a preset first loss function, the predicted value of the monitoring data for the next time in the historical time period, and the monitoring data for the next time in the historical time period; the parameters of the initial machine learning model are updated using the loss function value; if the updated initial machine learning model does not meet the preset training completion conditions, return to execute the step of selecting the target training sample currently used for training from each training sample in the training sample set.
[0074] S203: When the initial machine learning model meets the preset training completion conditions, the initial machine learning model that meets the training completion conditions is used as the trained machine learning model.
[0075] In this embodiment, the training completion condition may be that the loss function converges, the prediction accuracy of the initial machine learning model is greater than a preset accuracy threshold, or the number of training times is greater than a preset training number threshold, etc.
[0076] In an embodiment provided in the present application, based on the above scheme, optionally, the dynamic monitoring threshold range corresponding to the preset time is determined according to the monitoring data prediction value, such as Figure 3 As shown, including:
[0077] S301: Obtain error parameters corresponding to the monitoring data.
[0078] Optionally, the error parameter corresponding to the monitoring data may be calculated based on the monitoring data and a preset error ratio. The calculation method may be:
[0079] E=M×K 1
[0080] Among them, E is the error parameter, M is the monitoring data, k 1 is the error ratio.
[0081] In some embodiments, the error parameter of the monitoring data can also be calculated based on at least one target historical monitoring data and the monitoring prediction value corresponding to each target historical monitoring data; specifically, the absolute value of the difference between the target historical monitoring data and the monitoring prediction value corresponding to each target historical monitoring data can be calculated; the absolute values are averaged to obtain the error parameter of the monitoring data; wherein, the target historical monitoring data can be a target historical time in a historical period that matches a preset time; and the historical period is a period before the period in which the preset time is located.
[0082] For example, if the cycle length is 1 month and the preset time is May 1, the cycle of the preset time is from May 1 to May 31; the historical cycle can be March, January, etc., and the target historical time can be March 1, January 1, etc. If the cycle length is 1 year and the preset time is May 1, 2024, the target historical time can be May 1, 2023, May 1, 2022, etc.
[0083] S302: Determine a dynamic monitoring threshold range corresponding to the preset time according to the monitoring data prediction value and the error parameter.
[0084] In this embodiment, the monitoring data prediction value can be added with the error parameter to obtain the dynamic monitoring upper limit threshold of the dynamic monitoring threshold range; the monitoring data prediction value can be subtracted from the error parameter to obtain the dynamic monitoring lower limit threshold of the dynamic monitoring threshold range.
[0085] In an embodiment provided by the present application, based on the above solution, optionally, after determining that the monitoring data is abnormal, the method further includes:
[0086] Determining an abnormal level of the monitoring data;
[0087] Determining a display position of the monitoring data in a preset alarm page according to the abnormality level of the monitoring data;
[0088] The monitoring data is displayed at the display position in the alarm page.
[0089] In this embodiment, the difference value between the monitoring data and the predicted value of the monitoring data can be determined, and the difference value represents the degree of difference between the monitoring data and the predicted value of the monitoring data; the abnormality score of the monitoring data is determined according to the difference value and the priority score corresponding to the monitoring data; the abnormality level of the monitoring data is determined according to the abnormality score interval in which the abnormality score of the monitoring data is located; different abnormality scores correspond to different abnormality levels.
[0090] Optionally, the difference value between the monitoring data and the predicted value of the monitoring data may be an absolute value of the difference between the monitoring data and the predicted value of the monitoring data, or may be a ratio between the monitoring data and the predicted value of the monitoring data.
[0091] In this embodiment, different types of monitoring data correspond to the same or different priority scores. For example, the priority score of CPU utilization is higher than the priority score of interaction fluency.
[0092] In this embodiment, a preset configuration file can be queried according to the monitoring data to obtain the priority score corresponding to the monitoring data. The priority score of the monitoring data represents the importance of the monitoring data. The higher the priority score, the higher the correlation between the monitoring data and the safe and stable operation of the power business system. That is, if the monitoring data with a higher priority score is abnormal, it may cause the power business system to be unable to operate safely and stably.
[0093] Furthermore, the anomaly score of the monitoring data can be calculated based on the difference value and the priority score; when the difference value is the absolute value of the difference between the monitoring data and the predicted value of the monitoring data, the difference value and the priority score can be summed to obtain the anomaly score. When the difference value is the ratio between the monitoring data and the predicted value of the monitoring data, the anomaly score can be calculated in the following way:
[0094] Q=|1-K 2 |×P+R
[0095] Among them, Q is the anomaly score, K 2 is the ratio between the monitoring data and the predicted value of the monitoring data, P is the preset initial anomaly score, and R is the priority score.
[0096] In this embodiment, the abnormality level can be divided into multiple types, for example, it can be divided into important level and ordinary level; different levels correspond to different display bars in the alarm page, that is, the monitoring data can be displayed in the display position of the display bar corresponding to its abnormality level.
[0097] See also Figure 4 , is an example diagram of an alarm page provided in an embodiment of the present application. The alarm page may include a display bar for important alarm events and a display bar for common alarm events; the display bar for important alarm events may be used to display important-level monitoring data; the display bar for common alarm events may be used to display common-level monitoring data. For monitoring data at the same level, the monitoring data may be sorted and displayed in descending order according to the abnormality score of the monitoring data.
[0098] In an embodiment provided by the present application, based on the above solution, optionally, after displaying the monitoring data at the display position in the alarm page, the method further includes:
[0099] In case of receiving a processing operation for the monitoring data, determining a classification result of the monitoring data according to the processing operation, the classification result representing whether the abnormality determination is correct or incorrect;
[0100] The monitoring data is stored according to the classification result of the monitoring data.
[0101] In an embodiment of the present application, the user can perform a disposal operation through a selection control corresponding to the monitoring data in the alarm page.
[0102] Optionally, after storing the monitoring data according to the classification results of the monitoring data, the stored monitoring data can be used to update the parameters of the machine learning model to improve the prediction accuracy of the machine learning model.
[0103] and Figure 1 Corresponding to the method described above, the embodiment of the present application also provides a monitoring device for a business system, Figure 1 In the specific implementation of the method, the monitoring device of the business system provided in the embodiment of the present application can be applied to an electronic device, and its structural diagram is as follows Figure 5 As shown, specifically including:
[0104] The acquisition unit 501 is used to acquire monitoring data of the power business system within a preset time, and historical monitoring data of a time period before the preset time;
[0105] A prediction unit 502, configured to input historical monitoring data of a time period before the preset time into a pre-trained machine learning model to obtain a prediction result output by the machine learning model, wherein the prediction result includes a predicted value of the monitoring data of the power business system at the preset time;
[0106] A first determining unit 503 is used to determine the dynamic monitoring threshold range corresponding to the preset time according to the monitoring data prediction value;
[0107] A comparison unit 504, configured to compare the monitoring data with the dynamic monitoring threshold range;
[0108] The second determining unit 505 is configured to determine that the monitoring data is abnormal if the monitoring data is not within the dynamic monitoring threshold range.
[0109] In an embodiment provided in the present application, based on the above solution, optionally, the prediction unit 502 includes:
[0110] An acquisition subunit is used to acquire an initial machine learning model to be trained and a training data set; the training data set includes multiple training samples and a sample label of each training sample; each of the training samples includes monitoring data of a historical time period, and the sample label of each training sample includes monitoring data of the next time of the historical time period;
[0111] A training subunit, used to update the model parameters of the initial machine learning model using the training samples in the training data set to train the initial machine learning model;
[0112] The execution subunit is used to use the initial machine learning model that meets the preset training completion conditions as the trained machine learning model when the initial machine learning model meets the preset training completion conditions.
[0113] In an embodiment provided in the present application, based on the above solution, optionally, the first determining unit 503 includes:
[0114] An acquisition subunit, used to acquire error parameters corresponding to the monitoring data;
[0115] The first determination subunit is used to determine the dynamic monitoring threshold range corresponding to the preset time according to the monitoring data prediction value and the error parameter.
[0116] In an embodiment provided in the present application, based on the above solution, optionally, it further includes:
[0117] A third determining unit, configured to determine an abnormality level of the monitoring data;
[0118] a fourth determining unit, configured to determine a display position of the monitoring data in a preset alarm page according to an abnormality level of the monitoring data;
[0119] A display unit is used to display the monitoring data at the display position in the alarm page.
[0120] In an embodiment provided in the present application, based on the above solution, optionally, it further includes:
[0121] A receiving unit, configured to, upon receiving a processing operation for the monitoring data, determine a classification result of the monitoring data according to the processing operation, wherein the classification result represents whether the abnormality determination is correct or incorrect;
[0122] A storage unit is used to store the monitoring data according to the classification result of the monitoring data.
[0123] The specific principles and execution processes of each unit and module in the monitoring device of the business system disclosed in the above-mentioned embodiment of the present application are the same as those of the monitoring method of the business system disclosed in the above-mentioned embodiment of the present application. Please refer to the corresponding parts of the monitoring method of the business system provided in the above-mentioned embodiment of the present application, and will not be repeated here.
[0124] It should be noted that each embodiment in this specification is described in a progressive manner, and each embodiment focuses on the differences from other embodiments, and the same or similar parts between the embodiments can be referred to each other. For the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.
[0125] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.
[0126] For the convenience of description, the above device is described in various units according to their functions. Of course, when implementing the present application, the functions of each unit can be implemented in the same or multiple software and / or hardware.
[0127] It can be known from the description of the above implementation methods that those skilled in the art can clearly understand that the present application can be implemented by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solution of the present application can be essentially or partly contributed to the prior art in the form of a software product, which can be stored in a storage medium such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in the various embodiments of the present application or certain parts of the embodiments.
[0128] The above is a detailed introduction to a monitoring method for a business system provided by the present application. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea; at the same time, for general technical personnel in this field, according to the idea of the present application, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.
Claims
1. A monitoring method for a business system, characterized in that: include: Acquire monitoring data of the power business system within a preset time, as well as historical monitoring data of a time period before the preset time; Inputting historical monitoring data of a time period before the preset time into a pre-trained machine learning model to obtain a prediction result output by the machine learning model, wherein the prediction result includes a predicted value of the monitoring data of the power business system at the preset time; Determine the dynamic monitoring threshold range corresponding to the preset time according to the monitoring data prediction value; Comparing the monitoring data with the dynamic monitoring threshold range; When the monitoring data is not within the dynamic monitoring threshold range, it is determined that the monitoring data is abnormal.
2. The method according to claim 1, characterized in that The training process of the machine learning model includes: Acquire an initial machine learning model to be trained and a training data set; the training data set includes multiple training samples and a sample label of each training sample; each training sample includes monitoring data of a historical time period, and the sample label of each training sample includes monitoring data of the next time of the historical time period; Using the training samples in the training data set to update the model parameters of the initial machine learning model to train the initial machine learning model; When the initial machine learning model meets the preset training completion conditions, the initial machine learning model that meets the training completion conditions will be used as the trained machine learning model.
3. The method according to claim 1, characterized in that The determining, according to the monitoring data prediction value, a dynamic monitoring threshold range corresponding to the preset time includes: Obtaining error parameters corresponding to the monitoring data; The dynamic monitoring threshold range corresponding to the preset time is determined according to the monitoring data prediction value and the error parameter.
4. The method according to claim 1, characterized in that: After determining that the monitoring data is abnormal, the method further includes: Determining an abnormal level of the monitoring data; Determining a display position of the monitoring data in a preset alarm page according to the abnormality level of the monitoring data; The monitoring data is displayed at the display position in the alarm page.
5. The method according to claim 1, characterized in that After displaying the monitoring data at the display position on the alarm page, the method further includes: In case of receiving a processing operation for the monitoring data, determining a classification result of the monitoring data according to the processing operation, the classification result representing whether the abnormality determination is correct or incorrect; The monitoring data is stored according to the classification result of the monitoring data.
6. A monitoring device for a business system, characterized in that: include: An acquisition unit, used to acquire monitoring data of the power business system within a preset time, and historical monitoring data of a time period before the preset time; A prediction unit, used to input the historical monitoring data of the time period before the preset time into a pre-trained machine learning model to obtain a prediction result output by the machine learning model, wherein the prediction result includes a predicted value of the monitoring data of the power business system at the preset time; A first determining unit, configured to determine a dynamic monitoring threshold range corresponding to the preset time according to the monitoring data prediction value; A comparison unit, used for comparing the monitoring data with the dynamic monitoring threshold range; The second determining unit is configured to determine that the monitoring data is abnormal if the monitoring data is not within the dynamic monitoring threshold range.
7. The device according to claim 6, characterized in that The prediction unit comprises: An acquisition subunit is used to acquire an initial machine learning model to be trained and a training data set; the training data set includes multiple training samples and a sample label of each training sample; each of the training samples includes monitoring data of a historical time period, and the sample label of each training sample includes monitoring data of the next time of the historical time period; A training subunit, used to update the model parameters of the initial machine learning model using the training samples in the training data set to train the initial machine learning model; The execution subunit is used to use the initial machine learning model that meets the preset training completion conditions as the trained machine learning model when the initial machine learning model meets the preset training completion conditions.
8. The device according to claim 6, characterized in that The first determining unit includes: An acquisition subunit, used to acquire error parameters corresponding to the monitoring data; The first determination subunit is used to determine the dynamic monitoring threshold range corresponding to the preset time according to the monitoring data prediction value and the error parameter.
9. The device according to claim 6, characterized in that Also includes: A third determining unit, configured to determine an abnormality level of the monitoring data; a fourth determining unit, configured to determine a display position of the monitoring data in a preset alarm page according to an abnormality level of the monitoring data; A display unit is used to display the monitoring data at the display position in the alarm page.
10. The device according to claim 6, characterized in that Also includes: A receiving unit, configured to, upon receiving a processing operation for the monitoring data, determine a classification result of the monitoring data according to the processing operation, wherein the classification result represents whether the abnormality determination is correct or incorrect; A storage unit is used to store the monitoring data according to the classification result of the monitoring data.
