Scheduling automation system application access and performance tide analysis method

Abstract

The invention discloses a dispatching automation system application access and performance tide analysis method, and particularly relates to the field of power system automation, and the method comprises the steps: S1, access time sequence construction, S2, operation performance time sequence construction, S3, corresponding relation construction, S4, operation performance evaluation, S5, tide fluctuation analysis, S6, load fluctuation analysis, and S7, risk evaluation. The user access log and the application operation performance are monitored through the application access terminal and the performance sensing terminal, the access time sequence and the operation performance time sequence are constructed based on the monitoring result, the flexibility analysis of the electric power automation system is carried out, and the corresponding relation between the response time and the operation performance is constructed. And correlation comparison is carried out on the access time sequence, the response time and the performance stability parameters through the corresponding relation, so that the analysis on the tidal fluctuation and the load fluctuation is realized, and the application fluctuation condition of the tidal rule is better reflected and presented.

Description

Technical Field

[0001] The present invention relates to the technical field of power system automation, and more particularly to a method for application access and performance tidal analysis of a dispatching automation system. Background Art

[0002] Early power systems used centralized monitoring, such as SCADA systems, which relied on fixed rules and manual intervention and struggled to adapt to the dynamic demands of complex power grids. With the development of smart grids, power automation systems have gradually integrated distributed energy, IoT devices, and edge computing nodes. This has significantly increased system complexity, and system performance indicators have varied over time, closely related to application access and business processing, exhibiting a certain "tidal" pattern. Therefore, it is necessary to analyze the tidal patterns in the performance of power automation systems.

[0003] Existing power automation system application access and analysis methods include data collection, data prediction, threshold alarm, and visual interaction. Data collection uses sensors, smart meters and other devices to collect voltage, current, load and other data in real time, and combines monitoring and data collection, energy management and distribution management for centralized monitoring and control. Data prediction extracts patterns or predicts trends based on the collected data. Threshold alarm issues warnings based on fixed thresholds for predicted data. Fixed thresholds rely on manual experience to set parameters. Visual interaction visualizes data through terminals to support interactive queries and decision support.

[0004] Data analysis through existing power automation systems can identify risks and provide early warnings, but in actual use, it still has some shortcomings. First, the existing power automation system's business operations are complex, especially various business processing has its own time characteristics. User access to applications and the different operating performance of devices change over time. Based on this situation, it is necessary to dynamically process each application access and business. However, most existing analysis methods are limited to static thresholds and single-dimensional models for application access, resulting in the inability to capture the time-varying characteristics of business tides.

[0005] Second, in existing data prediction, device indicators and network data are usually stored in a dispersed manner. This easily leads to information silos, which is particularly harmful during fault analysis. This leads to inefficient fault location, thus affecting online services. Summary of the Invention

[0006] In view of this, an embodiment of the present invention provides a method for analyzing application access and performance tides in a scheduling automation system. User access logs and application operating performance are monitored through application access terminals and performance perception terminals, and access time series and operating performance time series are constructed based on the monitoring results. This enables analysis of tidal fluctuations and load fluctuations, while performing risk assessment of business scenarios and dynamically optimizing time windows based on the risk assessment results, effectively solving the problems raised in the background technology.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a method for application access and performance tidal analysis of a scheduling automation system, comprising an application access terminal, a performance perception terminal, and a control center, specifically comprising the following steps:

[0008] S1: Access time series construction: Obtain user access logs through the application access terminal. User access logs include user ID, access interface, and request frequency. This records the time when the user accesses the application and forms an access time series.

[0009] S2: Construction of running performance time series: Using the performance-aware terminal to perform real-time running performance detection while the application is running, and record the running performance detection time to form a running performance time series;

[0010] S3: Correspondence Construction: Map the access time series and the operation performance time series to establish the correspondence between response time and operation performance;

[0011] S4: Operational performance evaluation: CPU utilization, memory occupancy, and network throughput are sensed for the operational performance time series, and performance stability parameters are obtained for each operational performance detection time, thereby evaluating the operational performance;

[0012] S5: Tidal Fluctuation Analysis: Based on the corresponding relationship between access time series and response time and operation performance, the correlation between operation performance and response time is compared to analyze the tidal fluctuation coefficient;

[0013] S6: Load Fluctuation Analysis: The performance stability parameters of the operating performance time series are correlated with the access time series to form a correlation curve, thereby analyzing the application load fluctuation coefficient;

[0014] S7: Risk assessment: Perform risk assessment based on the tidal fluctuation coefficient and the applied load fluctuation coefficient, and optimize the time window based on the risk assessment results.

[0015] The technical effects and advantages of the present invention are as follows:

[0016] 1. The present invention monitors user access logs and application operating performance by using application access terminals and performance perception terminals, and constructs access time series and operating performance time series based on the monitoring results, thereby conducting flexibility analysis of the power automation system. It is not limited to analyzing application access based on static thresholds and single-dimensional models. On the one hand, it can meet the analysis needs to the greatest extent, and on the other hand, it can perform different time characteristic analyses on various types of business processing, which is conducive to more comprehensive reflection of the actual performance of operating performance during application access, greatly improving the reliability and practical value of risk assessment results.

[0017] 2. The present invention constructs a correspondence between response time and operating performance in the access time series and the operating performance time series, and correlates and compares the access time series with the response time and performance stability parameters through the correspondence, thereby realizing the analysis of tidal fluctuations and load fluctuations, directly binding user behavior with system resources, and quickly identifying response time, thereby ensuring the continuity of application access and use to a certain extent, and better reflecting the application fluctuations that show tidal patterns.

[0018] 3. The present invention evaluates the risk of business scenarios through tidal fluctuations and application load fluctuations, and dynamically optimizes the time window based on the risk assessment results. It avoids overlapping periods of high risk and high load through intersection calculation, reduces the probability of maintenance operations triggering an avalanche effect, and avoids affecting online business. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is a schematic diagram of the overall structure of the present invention.

[0020] Figure 2 Construct a flow chart for the correlation curve of the present invention.

[0021] Figure 3 Schematic diagram of the equipment connections used in the present invention. DETAILED DESCRIPTION

[0022] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0023] As attached Figure 1-Figure 3 A method for application access and performance tidal analysis of a scheduling automation system is shown, comprising an application access terminal, a performance perception terminal, and a control center.

[0024] In a more specific application of the present invention, the application access terminal is used to record the user access time when the user accesses the application; the performance perception terminal is used to perceive the operating performance data of the device when the user accesses it, specifically the perception of CPU utilization, memory occupancy and network throughput; the control center is used to perform analysis and control related parameters based on the application access terminal and the performance perception terminal.

[0025] The specific implementation of the present invention includes the following steps:

[0026] S1: Access time series construction: Obtain user access logs through the application access terminal. The user access logs include user ID, access interface, and request frequency, thereby recording the time when the user accesses the application to form an access time series.

[0027] S2: Construction of running performance time series: Use the performance perception terminal to perform real-time running performance detection when the application is running, and record the running performance detection time to form a running performance time series.

[0028] In this embodiment, it should be specifically explained that the operation performance test includes the detection of CPU utilization, memory occupancy and network throughput. The CPU utilization, memory occupancy and network throughput are used as operation performance tests because they directly reflect the calculation, storage and communication of the application. The user request is processed by the CPU. The higher the CPU utilization, the longer the task queue, which leads to longer response time; the storage resource situation is obtained by the memory occupancy. The higher the memory occupancy, the greater the response time fluctuation; the data transmission efficiency is reflected by the network throughput. The higher the network throughput, the smaller the packet loss and delay, and the shorter the response time.

[0029] It should be further explained that application access terminals and performance perception terminals need to maintain consistent timestamp accuracy when collecting user access logs and operating performance. Consistent timestamps ensure the temporal synchronization of user access logs and operating performance data. This means that the user access logs obtained at each time point can accurately correspond to the operating performance data at that time. Therefore, the response time of user access obtained through the user access log is also synchronized with the operating performance data in time, avoiding time deviations between the data. It also facilitates the mapping of subsequent access time series with operating performance time series.

[0030] S3: Correspondence construction: Map the access time series and the operation performance time series to establish the correspondence between response time and operation performance.

[0031] In this embodiment, it should be specifically explained that the corresponding relationship between response time and operating performance is established as follows:

[0032] A1: Number the corresponding performance test times in the performance time series in chronological order;

[0033] A2: Number the corresponding access times in the access time series in chronological order;

[0034] A3: Based on the one-to-one correspondence principle, the corresponding numbered running performance and access time are extracted from the running performance time series and access time series to form a mapping group;

[0035] A4: Obtain the response time corresponding to the running performance in each mapping group based on the running performance detection time and access time in each mapping group;

[0036] A5: A correspondence between response time and operating performance is established based on the response time corresponding to the operating performance in each mapping group.

[0037] It should be further explained that before establishing the correspondence between response time and operating performance, the access time series and the operating performance time series are mapped to align the time series data of different data sources. Through user access time and operating performance detection time, the correlation between user behavior and performance fluctuations can be quickly identified. When user access surges, the database query delay increases synchronously. In this case, the response time and operating performance constructed after mapping can correspond one to one, providing a basis for subsequent operating performance and load evaluation.

[0038] It needs to be explained that although the application access terminal and the performance perception terminal maintain consistent timestamp accuracy during collection, data collection is usually carried out in real time. Therefore, it is necessary to maintain a consistent collection frequency when collecting user access logs and running performance. At the same time, there is a certain time deviation in the formation process of the two time series. Since the interface calls between applications may cause delays, there will be errors in obtaining the response time corresponding to the running performance in each mapping group. Therefore, a suitable time window is introduced. For example, the time window can be 2 hours during the peak period of application usage, and the time window can be 0.1 hours during the low-peak period of application usage.

[0039] S4: Operational performance evaluation: The CPU utilization, memory occupancy, and network throughput of the operation performance time series are perceived separately to obtain the performance stability parameters of each operation performance detection time, thereby evaluating the operation performance.

[0040] In this embodiment, it should be specifically explained that the performance stability parameters specifically include a CPU fluctuation parameter, a memory fragmentation parameter, and a throughput stability parameter.

[0041] It should be further explained that CPU utilization is the percentage of CPU usage time in the running performance time series to the total application usage time. It is used to measure the occupancy of CPU resources. CPU utilization is perceived by the performance perception terminal, and the standard deviation of the CPU utilization of each running performance time series is calculated to obtain the CPU fluctuation parameter, which reflects the degree of fluctuation of CPU utilization.

[0042] The memory occupancy rate is the ratio of the memory space used in the running performance time series to the total memory space. The memory occupancy rate is perceived by the performance perception terminal, and the maximum continuous free blocks and the minimum continuous free blocks in each running performance time series are extracted. The maximum continuous free blocks are subtracted from the minimum continuous free blocks and then compared with the total memory amount to obtain the memory fragmentation parameter, which reflects the degree of memory fragmentation.

[0043] Network throughput is the amount of data successfully transmitted in the network per unit time of the operating performance time series. Network throughput is perceived by the performance perception terminal, and the average and standard deviation of the network throughput of each operating performance time series are calculated to obtain the average network throughput and network throughput standard deviation, respectively. The average network throughput and network throughput standard deviation are subtracted and compared with the average network throughput to obtain the throughput stability parameter, which reflects the stability of network throughput transmission.

[0044] It should be explained that the performance evaluation model is constructed based on CPU fluctuation parameters, memory fragmentation parameters, and throughput stability parameters. It is specifically expressed as follows:

[0045]

[0046] Where Pe represents the performance evaluation coefficient of each performance time series, Cb represents the CPU fluctuation parameter of each performance time series, Mb represents the memory fragmentation parameter of each performance time series, and Tb represents the throughput stability parameter of each performance time series. When the CPU fluctuation parameter is smaller, the memory fragmentation parameter is smaller, and the throughput stability parameter is larger, it means that the performance is better.

[0047] The present invention evaluates the operating performance from three aspects: CPU utilization, memory occupancy, and network throughput. This makes the performance evaluation more comprehensive and can more accurately reflect the overall effect of application access changing over time. Each aspect can capture different stability situations, thereby providing a more complete performance evaluation.

[0048] S5: Tidal Fluctuation Analysis: Based on the corresponding relationship between access time series and response time and operating performance, the correlation comparison between operating performance and response time is performed to analyze the tidal fluctuation coefficient.

[0049] In this embodiment, it should be specifically explained that the correlation and comparison process of the operating performance and response time is as follows:

[0050] Based on the access time series, a coordinate system is constructed with time as the horizontal axis and response time as the vertical axis, and several points are marked to form a response time change curve;

[0051] In the response time variation curve formed, the operation performance evaluation coefficient is used as the vertical coordinate, and the operation performance evaluation coefficient corresponding to each response time is marked according to the corresponding relationship between the response time and the operation performance, thereby forming an operation performance variation curve.

[0052] It should be added that drawing a horizontal axis and multiple different vertical axes in a coordinate system is usually used to represent the relationship between multiple different vertical axes under a specific horizontal axis. In this embodiment, since there is a corresponding relationship between response time and operating performance in the response time change curve constructed based on the access time series, an operating performance change curve can be constructed based on the response time change curve.

[0053] It should be further explained that the tidal fluctuation coefficient analysis is as follows:

[0054] B1: Obtain inflection points based on the response time change curve and mark the response time corresponding to each inflection point;

[0055] B2: Mark performance control points on the operating performance change curve based on the response time corresponding to each inflection point, obtain the tangent slope of each performance control point as the operating performance change rate of each performance control point, and extract the maximum operating performance change rate of each performance control point;

[0056] B3: The tidal fluctuation coefficient is obtained based on the operating performance change rate of each inflection point corresponding to the performance control point, which is specifically expressed as:

[0057]

[0058] Where TF represents the tidal fluctuation coefficient, r i Indicates the rate of change of the operating performance of the i-th inflection point corresponding to the performance control point, i = 1, 2, ..., n, r max It represents the maximum operating performance change rate of each performance control point. In this formula, when the response time changes, the smaller the operating performance change is, the smaller the tidal fluctuation coefficient is.

[0059] It should be explained that the inflection point is the point in the time series where the response time changes. When performing tidal fluctuation coefficient analysis, the point where the operating performance changes is selected on the operating performance change curve, and then the operating performance change rate corresponding to the point is obtained. The tidal fluctuation coefficient analysis is thus performed, thereby providing an accurate assessment of the power automation system under dynamic conditions, which can more flexibly reflect the operating performance during application access, and at the same time provide a basis for subsequent risk identification in different time periods; when the response time changes, the tidal fluctuation coefficient is further calculated by changing the operating performance. This is because the power automation system business is complex, and various business processing has its own time characteristics. User access to applications, interface calls between applications, and different operating performance indicators of equipment change over time, and are closely related to the access of each application and business processing process, showing a certain "tidal" law. Tidal fluctuation coefficient analysis can ensure that the fluctuation of operating performance is minimized under different response times.

[0060] S6: Load fluctuation analysis: The performance stability parameters of the operating performance time series and the access time series are combined to form a correlation curve, thereby analyzing the application load fluctuation coefficient.

[0061] In this embodiment, it should be specifically explained that the correlation curve is formed as follows: the response times in the access time series are arranged in descending order;

[0062] Based on the mapping group consisting of running performance and access time, the performance stability parameters of each response time corresponding to the running performance detection time are extracted. The performance stability parameters include CPU fluctuation parameters, memory fragmentation parameters, and throughput stability parameters.

[0063] A coordinate system is constructed with response time as the horizontal axis and CPU fluctuation parameters, memory fragmentation parameters, and throughput stability parameters as the vertical axes.

[0064] Points are marked in the coordinate system based on the order of response time and the performance stability parameters of each response time corresponding to the running performance test time, thereby forming a correlation curve between response time and CPU fluctuation, memory fragmentation, and throughput stability.

[0065] It should be further explained that the application load fluctuation coefficient analysis is as follows: based on the correlation curves, the inflection points within each correlation curve are obtained, and the slopes of each inflection point within each correlation curve are extracted and labeled as the CPU fluctuation change rate, memory fragmentation change rate, and throughput stability change rate, respectively.

[0066] Perform absolute value transformation on the CPU fluctuation change rate, memory fragmentation change rate, and throughput stability change rate, and extract the maximum and minimum change rates after the absolute value transformation.

[0067] Subtract the maximum change rate from the minimum change rate and compare it with the maximum change rate to obtain the change difference. The change difference specifically includes CPU fluctuation change difference, memory fragmentation change difference, and throughput stability change difference.

[0068] It should be explained that the maximum and minimum change rates are selected from the CPU fluctuations, memory fragmentation, and throughput stability change rates corresponding to each response time to calculate the change difference. This is because the maximum and minimum change rates can reflect the most severe changes, ensuring that the assessment of application load fluctuations includes the worst case scenario, thereby more accurately reflecting the stability of application access.

[0069] The CPU fluctuation variation, memory fragmentation variation, and throughput stability variation are calculated to obtain the application load fluctuation coefficient, which is specifically expressed as:

[0070]

[0071] Where LF represents the application load fluctuation coefficient, Ct represents the CPU fluctuation variation, Mt represents the memory fragmentation variation, and Tt represents the throughput stability variation. As can be seen from the above formula, by comparing each variation variation with the sum of the variation variations, the CPU fluctuation variation, memory fragmentation variation, and throughput stability variation are given equal weight when analyzing the application load fluctuation coefficient, thus avoiding an excessive impact on the results due to an excessively large numerical range of a single indicator.

[0072] It should be explained that high CPU fluctuation variability may be due to the sudden increase or decrease in complex queries, resulting in unstable demand for CPU resources, which in turn affects the load stability of the entire application; when the memory fragmentation variability is large, it indicates that the degree of memory fragmentation is unstable, which leads to unstable application response time and increased application load fluctuations; high throughput stability variability affects the data interaction between the application and the power automation system, and thus affects the overall performance and load stability of the application, resulting in an increase in the application load fluctuation coefficient; high CPU fluctuations may cause the application to process data at an unstable speed, which in turn affects the stability of network throughput and memory fragmentation changes. Therefore, when analyzing the application load fluctuation coefficient, it is necessary to comprehensively consider the impact of these three variabilities to accurately reflect the actual fluctuations in the application load.

[0073] S7: Risk assessment: Perform risk assessment based on the tidal fluctuation coefficient and the applied load fluctuation coefficient, and optimize the time window based on the risk assessment results.

[0074] In this embodiment, it should be specifically explained that the risk assessment is as follows:

[0075] Based on the square sum calculation of the tidal fluctuation coefficient and the applied load fluctuation coefficient, a comprehensive risk assessment index is obtained, which is specifically expressed as:

[0076]

[0077] RE represents the comprehensive risk assessment index, TF represents the tidal fluctuation coefficient, and LF represents the application load fluctuation coefficient. The greater the change in operating performance over time and the greater the application load fluctuation, the higher the business scenario risk when using the application and the worse the application access performance.

[0078] Based on the distribution of historical data, the risk threshold is dynamically set using the quantile method. For example, the risk threshold is 95% of the historical data. The comprehensive risk assessment index is compared with the risk threshold. If the comprehensive risk assessment index is greater than or equal to the risk threshold, it means that the business scenario risk is high when the application is used. In this case, this business scenario is marked as high risk. At this time, automatic scaling is enabled through the control center. If the comprehensive risk assessment index is less than the risk threshold, it means that the business scenario risk is low when the application is used, and this business scenario is marked as low risk.

[0079] It should be further explained that the time window optimization is as follows: business scenarios marked as high-risk and low-risk are set as the peak and low-peak periods of application usage respectively;

[0080] Statistics are collected for the response time corresponding to when it is marked as high risk, and the high-risk operating performance monitoring time is extracted based on the corresponding relationship between the response time and the operating performance. The high-risk operating performance monitoring time is intersected with the time window of the current application usage peak to obtain a new time interval, and the new time interval is updated as a new time window.

[0081] The present invention establishes a correspondence between response time and operating performance in access time series and operating performance time series, and correlates and compares access time series with response time and performance stability parameters through the correspondence, thereby realizing the analysis of tidal fluctuations and load fluctuations, directly binding user behavior with system resources, quickly identifying response time, and evaluating the risk of business scenarios through tidal fluctuations and application load fluctuations. The time window is dynamically optimized according to the risk assessment results, and overlapping periods of high risk and high load are avoided through intersection calculation, thereby reducing the probability of maintenance operations causing an avalanche effect and avoiding impacts on online business.

[0082] Secondly: The drawings of the embodiments disclosed in the present invention only involve structures related to the embodiments disclosed in the present invention. Other structures may refer to conventional designs. The same embodiment and different embodiments of the present invention may be combined with each other without conflict.

[0083] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for analyzing application access and performance tidal trends in a dispatching automation system, characterized in that: include: S1: Access time series construction: Obtain user access logs through the application access terminal. User access logs include user ID, access interface, and request frequency. This records the time when the user accesses the application and forms an access time series. S2: Construction of running performance time series: Using the performance-aware terminal to perform real-time running performance detection while the application is running, and record the running performance detection time to form a running performance time series; S3: Correspondence Construction: Map the access time series and the operation performance time series to establish the correspondence between response time and operation performance; S4: Operational performance evaluation: CPU utilization, memory occupancy, and network throughput are sensed for the operational performance time series, and performance stability parameters are obtained for each operational performance detection time, thereby evaluating the operational performance; S5: Tidal Fluctuation Analysis: Based on the corresponding relationship between access time series and response time and operation performance, the correlation between operation performance and response time is compared to analyze the tidal fluctuation coefficient; S6: Load Fluctuation Analysis: The performance stability parameters of the operating performance time series are correlated with the access time series to form a correlation curve, thereby analyzing the application load fluctuation coefficient; S7: Risk assessment: Perform risk assessment based on the tidal fluctuation coefficient and the applied load fluctuation coefficient, and optimize the time window based on the risk assessment results.

2. A method for analyzing application access and performance tidal trends in a dispatching automation system according to claim 1, characterized in that: The corresponding relationship between the response time and the operating performance is established as follows: A1: Number the corresponding performance test times in the performance time series in chronological order; A2: Number the corresponding access times in the access time series in chronological order; A3: Based on the one-to-one correspondence principle, the corresponding numbered running performance and access time are extracted from the running performance time series and access time series to form a mapping group; A4: Obtain the response time corresponding to the running performance in each mapping group based on the running performance detection time and access time in each mapping group; A5: A correspondence between response time and operating performance is established based on the response time corresponding to the operating performance in each mapping group.

3. A method for analyzing application access and performance tides in a dispatching automation system according to claim 1, characterized in that: The performance stability parameters specifically include CPU fluctuation parameters, memory fragmentation parameters, and throughput stability parameters.

4. A method for analyzing application access and performance tides in a dispatching automation system according to claim 3, characterized in that: The CPU fluctuation parameter is obtained by sensing the CPU utilization through the performance perception terminal and calculating the standard deviation of the CPU utilization of each running performance time series; The memory fragmentation parameter is obtained by sensing the memory occupancy rate through the performance perception terminal, extracting the maximum continuous free block and the minimum continuous free block in each running performance time series, subtracting the maximum continuous free block from the minimum continuous free block and comparing it with the total memory capacity; The throughput stability parameter is obtained by sensing the network throughput through the performance perception terminal, and calculating the average and standard deviation of the network throughput of each running performance time series to obtain the average network throughput and the standard deviation of the network throughput. The average network throughput and the standard deviation of the network throughput are subtracted and compared with the average network throughput.

5. The method for analyzing application access and performance tide of a dispatching automation system according to claim 1, characterized in that: The correlation and comparison process of the operating performance and response time is as follows: Based on the access time series, a coordinate system is constructed with time as the horizontal axis and response time as the vertical axis, and several points are marked to form a response time change curve; In the response time variation curve formed, the operation performance evaluation coefficient is used as the vertical coordinate, and the operation performance evaluation coefficient corresponding to each response time is marked according to the corresponding relationship between the response time and the operation performance, thereby forming an operation performance variation curve.

6. A method for analyzing application access and performance tides in a dispatching automation system according to claim 1, characterized in that: The tidal fluctuation coefficient is analyzed as follows: B1: Obtain inflection points based on the response time change curve and mark the response time corresponding to each inflection point; B2: Mark performance control points on the operating performance change curve based on the response time corresponding to each inflection point, obtain the tangent slope of each performance control point as the operating performance change rate of each performance control point, and extract the maximum operating performance change rate of each performance control point; B3: The tidal fluctuation coefficient is obtained based on the operating performance change rate of the performance control point corresponding to each inflection point.

7. The method for analyzing application access and performance tide of a dispatching automation system according to claim 1, characterized in that: The application load fluctuation coefficient is analyzed as follows: the inflection points in each correlation curve are obtained according to the correlation curve, and the slopes of the inflection points in each correlation curve are extracted, which are marked as the CPU fluctuation change rate, the memory fragmentation change rate, and the throughput stability change rate respectively; Perform absolute value transformation on the CPU fluctuation change rate, memory fragmentation change rate, and throughput stability change rate, and extract the maximum and minimum change rates after the absolute value transformation. Subtract the maximum change rate from the minimum change rate and compare it with the maximum change rate to obtain the change difference. The change difference specifically includes CPU fluctuation change difference, memory fragmentation change difference, and throughput stability change difference. The CPU fluctuation variation, memory fragmentation variation, and throughput stability variation are calculated to obtain the application load fluctuation coefficient, which is specifically expressed as: LF represents the application load fluctuation coefficient, Ct represents the CPU fluctuation variance, Mt represents the memory fragmentation variance, and Tt represents the throughput stability variance.

8. The method for analyzing application access and performance tide of a dispatching automation system according to claim 1, characterized in that: The time window optimization is as follows: business scenarios marked as high risk and low risk are set as application usage peak period and low peak period respectively; Statistics are collected for the response time corresponding to when it is marked as high risk, and the high-risk operating performance monitoring time is extracted based on the corresponding relationship between the response time and the operating performance. The high-risk operating performance monitoring time is intersected with the time window of the current application usage peak to obtain a new time interval, and the new time interval is updated as a new time window.

Images