Real-time processing method for collected system state data

By determining the data acquisition period and processing singular points, smoothing filtering and eliminating integral drift, the error problem of sensor data processing is solved, the accuracy and reliability of system status data is achieved, and effective data support is provided for the control system.

CN118964856APending Publication Date: 2024-11-15中国人民解放军32272部队41分队
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411197369.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-29
Publication Date
2024-11-15

Smart Images

  • Figure CN118964856A_ABST
    Figure CN118964856A_ABST
Patent Text Reader

Abstract

The invention discloses a real-time processing method for collected system state data, and relates to the technical field of data signal processing, and the method specifically comprises the following steps: 1, determining a control system state collection parameter according to the requirements of a controlled mechanism, a control quantity, real-time performance and the like; 2, singular points in the real-time collected data are eliminated through a reference historical data set; step 3, performing smooth filtering processing on the data after singular point elimination through a smooth attention data set; and step 4, performing integral drift elimination processing on the data after smooth filtering through the integral attention data set. According to the method, the priorities of multiple maintenance tasks received by the equipment maintenance institution can be reasonably and effectively ranked, a feasible task maintenance sequence is given, and effective theoretical support and technical support can be provided for optimal configuration, scheduling application and the like of guarantee resources such as maintenance personnel, machine tool equipment and maintenance equipment of the equipment maintenance institution.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of data signal processing, and in particular to a real-time processing method for collected system status data. Background Art

[0002] Real-time control systems, such as state detection systems, signal processing systems, and operation control systems, all need to collect state data through state sensors arranged at key positions in the system. After preliminary processing of the data, the control quantity is obtained through professional control calculations, and then transmitted to the actuator to complete closed-loop real-time control according to certain control rules and strategies. It has a wide range of applications in modern industrial, engineering, and mechanical fields.

[0003] In the operation and execution of the control system, data collection and processing are its most basic and fundamental work. The real-time data collected by the sensor will be affected by random signal errors and inherent detection errors, so that the collected data cannot be directly used for calculation and must be processed. At present, there are two main data processing methods: hardware filtering and software program processing. For hardware filtering, although the structure is simple, the function is relatively single and solidified, and it cannot be adjusted at any time, and the accuracy of data processing is not high. The data processing of the software program is to process the data in real time through the software program in the control system. On the one hand, the processing method can be adjusted in real time according to the needs, and on the other hand, it can save the hardware system layout resources and effectively improve the miniaturization and integration characteristics of the control system. Attention should be paid to the complexity of the software algorithm to ensure the real-time performance of data processing.

[0004] At present, the processing of collected system status data usually relies mainly on the experience of system designers to make judgments and analyses. There is no basically unified and consistent data processing method between different control systems and control quantities. Therefore, the present invention provides a general real-time processing method for sensor collected data based on the real-time control system, providing effective data processing methods and technical support for control system design and development engineers. Summary of the invention

[0005] The present invention provides a real-time processing method for collected system status data, which solves the problem that the data collected by the sensor disclosed in the background technology cannot be directly used for algorithm design and system control.

[0006] The present invention provides the following technical solution: a real-time processing method for collected system status data, comprising the following steps: Step 1: Determine the data acquisition cycle frequency, data signal type and data unit scale of the control system according to the controlled mechanism, control quantity and real-time requirements; The specific operation is: according to the requirements of the controlled mechanism, control quantity and real-time performance, clarify the indicators and model requirements of the sensors for collecting various control parameters of the control system, and determine the data collection cycle of the control system by developing the control board program. , data collection frequency , control system operation clock step , the signal response time limit of the controlled agency ; Step 2: Determine the reference historical data set according to the signal response requirements and the data collection cycle, identify the singular points of the system status data in real time according to the rolling reference historical data set, and remove them; The specific operation is: determine the range of system state data, and select a suitable method for eliminating singular points of system state data according to the determined range; Step 3: Determine a smoothed data set of interest based on signal response requirements and data acquisition cycle, and perform smoothing and filtering processing on the system state in real time based on the rolling smoothed data set of interest; The specific operations are: 1. Construct a smooth attention data set; 2. Use the following formula to smooth the system status data: ; in: represents the important coefficient of the element in the smoothing attention data set, and ; Step 4: for the system state quantity to be integrated, determine the integral focus data set according to the signal response requirements and the data acquisition cycle, and perform integral drift elimination processing on the system state in real time according to the rolling integral focus data set; The specific operations are as follows: 1. Constructing a points attention data set; 2. Perform curve fitting on the data in the integral focus data set in order to eliminate the integral drift of the system state data. The curve used for fitting is: ; in: It represents the first data, represents the characteristic parameters of the curve calculated by least squares fitting; 3. The effective system status data after eliminating the integral drift error is: ; in: Indicates the collected System status data.

[0007] Preferably, the control system operation clock step refers to the time required to run each basic operation during the system program calculation process. The basic operation includes all operation forms commonly used in computing software. To facilitate statistical calculations, different types of operations are defined as the same operation duration.

[0008] Preferably, the signal response time limit of the controlled mechanism is Includes upper limit on signal response time of the accused institution and the lower limit of the signal response time of the controlled organization ; is the longest time interval of the control quantity, is the shortest time interval of the control quantity.

[0009] Preferably, the step 2 has three types of elimination processing methods for the singular points of the system state data: 1. When the system status data has a certain valid range, that is, the data has a clear upper and lower limit, then: ; in: Indicates the collected system status data, Indicates the upper limit of system status data, Indicates the lower limit of system status data; 2. When the system status data has a fuzzy effective range, that is, there is no clear upper and lower limits, but there is a certain reference range in terms of the actual physical meaning of the data, then: ; in: Indicates the physical upper limit of the system status data. Indicates the physical lower limit of system status data; 3. When the system status data has no valid range, that is, the data may increase or decrease over time, or change continuously over a large range, it is necessary to build a reference historical data set, remove singular points based on the data change trend in the recent period of time, and take the most recently collected system status data. , amount of data The following constraints must be met: ; in, represents the reference historical dataset, Indicates the first elements, the data in the reference historical data set is updated in real time as the data is collected. After the reference historical data set is determined, there is a singular point elimination method: ; in, Indicates the system status data obtained from the last sampling.

[0010] Preferably, in step 3, when constructing the smoothed attention data set, the most recently collected system status data is taken , amount of data The following constraints are met: ; in, represents a smooth attention dataset, represents the i-th element in the smoothing attention dataset.

[0011] Preferably, when constructing the integral focus data set, the most recent collected and integrated system status data is taken. , amount of data The following constraints need to be met: ; in, represents the integral attention dataset, Indicates the first elements.

[0012] Preferably, in step 4, the step of obtaining the curve used for fitting is: 21. Set the curve used for fitting as: ; in: It represents the first data, and is the characteristic parameter of the curve; when and When the following requirements are met, the sample points are closest to the curve used for fitting, that is: ; in: represents the characteristic parameters of the curve calculated by least squares fitting; 22. Order ,make right and Find the partial derivative and solve it jointly to get: ; ; 2.3. The curve used for fitting is: .

[0013] Preferably, the data in the integral focus data set is updated in real time as the data is collected; and the data in the smoothing focus data set is updated in real time as the data is collected.

[0014] Compared with the prior art, the present invention has the following beneficial effects: The present invention processes the collected system status data in a standardized, effective and practical manner from four aspects. By determining the basic information such as the collected status data type, collection frequency and data scale of the control system, the feasible range and effective range of the system can be standardized and defined. By analyzing the system status data with different range characteristics, the singular points in the data are eliminated in a targeted manner to improve the validity and reliability of the data. On this basis, the smoothing filtering of the system status data is realized through the program formula to remove the noise and burrs in the data and improve the accuracy and precision of the data. For the system status data that cannot be directly collected by the sensor and requires data integration operation, the least squares method is used to eliminate the integral drift of the data, and effective system state estimation data can be obtained to improve the accuracy and validity of the data. The system status data processed by the above method can be directly used for the next step of system control, decision-making, classification and identification, and can provide the system with an effective, accurate and reliable data foundation and data support.

[0015] 2. The real-time processing method of the collected system status data can reasonably and effectively sort the priorities of multiple maintenance tasks received by equipment maintenance organizations, and provide a feasible task maintenance sequence, which can provide effective theoretical and technical support for the optimal allocation and scheduling of maintenance personnel, machinery and equipment, maintenance equipment and other support resources of equipment maintenance organizations. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 The present invention is a flowchart of a method for real-time processing of collected system status data.

[0017] Figure 2 This is the data diagram of the vehicle body vertical vibration acceleration state collected by the sensor.

[0018] Figure 3 This is the data diagram of the vertical vibration acceleration state of the vehicle body after eliminating the singular points.

[0019] Figure 4 This is the data diagram of the vehicle body vertical vibration acceleration state after filtering.

[0020] Figure 5 This is the vehicle body vertical vibration acceleration state data (slice capture) after filtering.

[0021] Figure 6 This is the data diagram of the vehicle body vertical vibration velocity state after integration.

[0022] Figure 7 This is the data diagram of the vehicle body vertical vibration velocity state after the integral drift is eliminated.

[0023] Figure 8 This is a comparison diagram of the vehicle body vertical vibration velocity state data (slice capture). DETAILED DESCRIPTION

[0024] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. 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 creative work are within the scope of protection of the present invention.

[0025] The present invention provides a real-time processing method for collected system status data, comprising the following steps: Step 1: Determine the data acquisition cycle frequency, data signal type and data unit scale of the control system according to the controlled mechanism, control quantity and real-time requirements; The specific operation is: according to the requirements of the controlled mechanism, control quantity and real-time performance, clarify the indicators and model requirements of the sensors for collecting various control parameters of the control system, and determine the data collection cycle of the control system by developing the control board program. , data collection frequency , control system operation clock step , the signal response time limit of the controlled agency ; Among them: the control system operation clock step refers to the time required to run each basic operation during the system program calculation process. Basic operations include all commonly used operation forms of computing software such as arithmetic operations, set operations, algebraic operations, and bit operations. For the convenience of statistical calculation, different types of operations are set to the same operation time.

[0026] The signal response time limit of the controlled mechanism is also the frequency interval requirement of the control system control quantity output, which usually has an upper limit and a lower limit. It is the longest time interval of the controlled quantity. When the control interval is greater than the upper limit, that is, the interval from the acquisition of the system status signal to the output of the control signal is too long, it will cause the time phase of the controlled quantity to be disordered, and the controlled mechanism cannot be effectively controlled, and may even have a negative effect. The lower limit of the signal response time of the controlled mechanism It is the shortest time interval of the control quantity. When the control quantity interval is less than the lower limit, that is, the interval from the acquisition of the system state signal to the output of the control signal is too short, the control quantity actuator will not be able to complete the response execution of the control quantity, and it will not be able to effectively control the controlled mechanism. Therefore, when designing the control algorithm of the control system, it is necessary to focus on the size of the algorithm calculation in order to reasonably adjust the output interval of the control quantity.

[0027] Step 2: Determine the reference historical data set according to the signal response requirements and the data collection cycle, identify the singular points of the system status data in real time according to the rolling reference historical data set, and remove them; The specific operation is: determine the range of system state data, and select a suitable method for eliminating singular points of system state data according to the determined range; In this step, there are three types of elimination processing methods for the singular points of the system state data: 1. When the system state data has a certain effective range, that is, the data has a clear upper and lower limit, then: ; in: Indicates the collected system status data, Indicates the upper limit of system status data, Indicates the lower limit of system status data; 2. When the system status data has a fuzzy effective range, that is, there is no clear upper and lower limits, but there is a certain reference range in terms of the actual physical meaning of the data, then: ; in: Indicates the physical upper limit of the system status data. Indicates the physical lower limit of system status data; 3. When the system status data has no valid range, that is, the data may increase or decrease over time, or change continuously over a large range, it is necessary to build a reference historical data set, remove singular points based on the data change trend in the recent period of time, and take the most recently collected system status data. , amount of data The following constraints must be met: ; in, represents the reference historical dataset, Indicates the first elements, the data in the reference historical data set is updated in real time as the data is collected. After the reference historical data set is determined, there is a singular point elimination method: ; in, Indicates the system status data obtained from the last sampling.

[0028] Step 3: Determine a smoothed data set of interest based on signal response requirements and data acquisition cycle, and perform smoothing and filtering processing on the system state in real time based on the rolling smoothed data set of interest; The specific operations are as follows: 1. Construct a smoothing data set. The size of the data set has a great influence on the filtering effect of the collected data. If the data set is too small, the noise and burrs in the collected system status data cannot be effectively filtered out; if the data set is too large, it may cause the status data to be over-smoothed and distorted, and lose effective data information. Take the most recent collected system status data , amount of data The following constraints are met: ; in, represents a smooth attention dataset, represents the i-th element in the smoothing attention dataset; 2. Use the following formula to smooth the system status data: ; in: represents the important coefficient of the element in the smoothing attention data set, and The data set size under the above constraints is usually a large range, and it is necessary to debug the effective data size in a targeted manner for different state data. The data in the smoothing focus data set is updated in real time as the data is collected. After the smoothing focus data set is determined, the system state data is smoothed according to the above smoothing filter formula, and usually the more recently collected data is, the greater its importance coefficient is.

[0029] Step 4: for the system state quantity to be integrated, determine the integral focus data set according to the signal response requirements and the data acquisition cycle, and perform integral drift elimination processing on the system state in real time according to the rolling integral focus data set; The specific operations are as follows: 1. Construct an integral focus data set (a data set with drift after the system state collected by the sensor is integrated). The size of the data set also has a great impact on the system state estimation effect of the data. If the data set is too small, the drift trend and size after the data integration cannot be effectively perceived; if the data set is too large, it may not meet the real-time requirements of signal control due to excessive calculation amount and too long calculation time. Take the most recent collected and integrated system state data , amount of data The following constraints need to be met: ; in, represents the integral attention dataset, Indicates the first elements. Similarly, the data set size under the above constraints is usually a large range, and it is necessary to debug the effective data size in a targeted manner for different state data. The data in the integral focus data set is updated in real time as the data is collected. After the integral focus data set is determined, there is a method for eliminating the integral drift of the system state data.

[0030] 2. Perform curve fitting on the data in the integral focus data set in order to eliminate the integral drift of the system state data. The curve used for fitting is: ; in: It represents the first data, represents the characteristic parameters of the curve calculated by least squares fitting; The steps for obtaining the curve used in the above fitting are: 21. Set the curve used for fitting as: ; in: It represents the first data, and is the characteristic parameter of the curve; when and When the following requirements are met, the sample points are closest to the curve used for fitting, that is: ; in: represents the characteristic parameters of the curve calculated by least squares fitting; 22. Order ,make right and Find the partial derivative and solve it jointly to get: ; ; 2.3. The curve used for fitting is: .

[0031] The curve used in the above fitting is At 12:00, Each integral focuses on the integral drift trend curve obtained by fitting historical data; 3. The effective system status data after eliminating the integral drift error is: ; in: Indicates the collected System status data.

[0032] Example The present invention provides a method for real-time processing of collected system status data. Figure 1 As shown, it mainly includes the following steps: Step 1: Assume that a real-time control system is considered, which needs to perform semi-active vibration control on the vehicle suspension system to reduce the vertical vibration acceleration at the vehicle seat and improve the stability of vehicle driving. This real-time vibration control system of the vehicle suspension has two acceleration sensors and one displacement sensor, which are arranged at the upper end, lower end and side by side of the suspension damper, respectively, and are mainly used to collect the vertical vibration acceleration of the vehicle body and wheels and the relative displacement state data between them.

[0033] The frequency range of vehicle suspension vibration is 0.2Hz to 30Hz, so it can be determined that the data acquisition frequency should be greater than 60Hz, which can be set as =100Hz, that is, the data acquisition cycle is =0.01s. Through customized development, it is known that the control system operation clock step is Considering the reaction time of the semi-active control system based on the damper, the lower limit of the signal response time of the controlled mechanism can be determined =0.002s; Considering the upper frequency limit of vehicle suspension vibration, the upper limit of the signal response time of the controlled mechanism can be determined =0.01s. From the above data, we can see that the current system data acquisition cycle is exactly the same as the upper limit of the signal response time of the controlled mechanism, that is, taking the acquisition cycle as the control signal output cycle can meet the vehicle suspension vibration control requirements.

[0034] Step 2: Give the data collected by the acceleration sensor at the upper end of the damper in the suspension, that is, the vertical vibration acceleration state data of the vehicle body, such as Figure 2 shown.

[0035] The vertical vibration acceleration state data of the vehicle body does not have a definite effective range when driving on a grade highway, that is, there is no clear upper and lower limit. However, based on experience, when driving on a Class B highway with good road roughness characteristics, the maximum vertical vibration acceleration of the vehicle body usually does not exceed 5m / s 2 , usually the effective value is 2m / s 2 If the vehicle body vertical vibration acceleration state data is within the range, the method for eliminating singular points is: ; Then the vertical vibration acceleration state data of the vehicle body after removing the singular point can be obtained as follows: Figure 3 shown.

[0036] Since the original collected state data has good uniformity, it can be seen that the data of the three singular points are not within the effective range of the collected data. However, in the real-time acquisition control system, it is still necessary to go through a filtering process to smooth it so that the collected system state data can be used directly.

[0037] Step 3: After the singular points are eliminated, the collected system status data needs to be smoothed and filtered to remove the influence of noise and burrs.

[0038] Assume that the amount of data in the smoothing attention data set is 5, that is , satisfying the corresponding constraints: ; Assume that the important coefficients of the elements in the smoothing attention data set are all 1, that is, , indicating that the importance of each data in the smoothing data set is consistent, which is a standard smoothing operation. The smoothing filter formula is: ; The vehicle body vertical vibration acceleration state data after filtering can be obtained as follows: Figure 4 As shown in the figure, it can be seen that the data with obvious burrs after removing the singular points are effectively smoothed, and the data are all within the effective range. Furthermore, the state data from 40 s to 42 s is sliced ​​and intercepted. It can be seen that the filtered data can fully characterize the changing trend of the system state and eliminate invalid noise and spikes.

[0039] Step 4: When considering the vehicle suspension vibration control, the vertical vibration velocity of the vehicle body is required as the control signal input, and the control quantity is obtained and output to the damper according to the semi-active control algorithm. The vertical vibration velocity of the vehicle body cannot usually be effectively collected by sensors and can only be obtained by state estimation and other calculations. Here, the collected acceleration state data is integrated to obtain the integrated vertical vibration velocity state data of the vehicle body, such as Figure 6 As shown. It can be seen that the vehicle body vertical vibration velocity state data will produce approximately linear drift as time increases, causing the velocity state data to gradually deviate from the effective range as time increases. This system state data after direct integration is invalid and cannot be used. Here, the system state estimation method based on least squares provided by the present invention is used to fit the data drift trend and reduce it, thereby eliminating the influence of integral drift.

[0040] Assume that the amount of data in the integral attention data set is , the corresponding constraints are as follows: ; In real time, 2000 data sets of integral attention are used as the data basis, and the drift trend is fitted and eliminated by the least square method to obtain the vertical vibration velocity state data of the vehicle body after the integral drift is eliminated, such as Figure 7 It can be seen that after the integral drift is eliminated, the vehicle vertical vibration velocity state data fluctuates around 0, meeting the requirements of data randomness and uniformity, which illustrates the effectiveness and feasibility of using the least squares method to eliminate integral drift.

[0041] Furthermore, the vertical vibration acceleration data of the vehicle body can be obtained by differential calculation of the data, and the acceleration data obtained by back calculation after processing can be compared with the original acceleration data after singular point elimination and smooth filtering, such as Figure 8 As shown in the figure, the state data from 40s to 42s are sliced ​​and it can be seen that the two acceleration state data are basically consistent, indicating that the speed state data after the integral drift is eliminated does not affect or change the basic information of the original acceleration state data. It also illustrates the feasibility and reliability of using the least squares method to eliminate the integral drift.

[0042] To sum up: the present invention adopts the above-mentioned real-time processing method of the collected system status data, based on the signal processing requirements, processing rules, and processing methods, to perform real-time and effective processing on the collected system status data, eliminate singular points in the status data, perform smoothing filtering to eliminate noise, spikes and burrs, and use the least squares fitting method to solve the problem of state data drift after integration, so as to provide effective and feasible system status data for the control system to be examined, and provide effective technical support and data support for the system control, decision-making, classification and identification.

[0043] Finally, it should be noted that the above is only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the embodiments, those skilled in the art should understand that they can still modify or replace the technical solution of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solution to deviate from the spirit and scope of the technical solution of the present invention.

Claims

1. A method for real-time processing of collected system status data, characterized in that: The following steps are involved: Step 1: Determine the data acquisition cycle frequency, data signal type and data unit scale of the control system according to the controlled mechanism, control quantity and real-time requirements; The specific operation is: according to the requirements of the controlled mechanism, control quantity and real-time performance, clarify the indicators and model requirements of the sensors for collecting various control parameters of the control system, and determine the data collection cycle of the control system by developing the control board program. , data collection frequency , control system operation clock step , the signal response time limit of the controlled agency ; Step 2: Determine the reference historical data set according to the signal response requirements and the data collection cycle, identify the singular points of the system status data in real time according to the rolling reference historical data set, and remove them; The specific operation is: determine the range of system state data, and select a suitable method for eliminating singular points of system state data according to the determined range; Step 3: Determine a smoothed data set of interest based on signal response requirements and data acquisition cycle, and perform smoothing and filtering processing on the system state in real time based on the rolling smoothed data set of interest; The specific operations are:

1. Construct a smooth attention data set; 2. Use the following formula to smooth the system status data: ; in: represents the important coefficient of the element in the smoothing attention data set, and ; Step 4: for the system state quantity to be integrated, determine the integral focus data set according to the signal response requirements and the data acquisition cycle, and perform integral drift elimination processing on the system state in real time according to the rolling integral focus data set; The specific operations are as follows:

1. Constructing a points attention data set; 2. Perform curve fitting on the data in the integral focus data set in order to eliminate the integral drift of the system state data. The curve used for fitting is: ; in: It represents the first data, represents the characteristic parameters of the curve calculated by least squares fitting; 3. The effective system status data after eliminating the integral drift error is: ; in: Indicates the collected System status data.

2. The method for real-time processing of collected system status data according to claim 1, characterized in that: The control system operation clock step refers to the time required to run each basic operation during the system program calculation process. The basic operation includes all operation forms commonly used by the calculation software. To facilitate statistical calculations, different types of operations are set to the same operation time.

3. The real-time processing method of the collected system status data according to claim 1 is characterized in that: The signal response time limit of the controlled mechanism Includes upper limit on signal response time of the accused institution and the lower limit of the signal response time of the controlled organization ; is the longest time interval of the control quantity, is the shortest time interval of the control quantity.

4. The method for real-time processing of collected system status data according to claim 1, characterized in that: The step 2 has three types of elimination processing methods for the singular points of the system state data:

1. When the system status data has a certain valid range, that is, the data has a clear upper and lower limit, then: ; in: Indicates the collected system status data. Indicates the upper limit of system status data, Indicates the lower limit of system status data; 2. When the system status data has a fuzzy effective range, that is, there is no clear upper and lower limits, but there is a certain reference range in terms of the actual physical meaning of the data, then: ; in: Indicates the physical upper limit of the system status data. Indicates the physical lower limit of system status data; 3. When the system status data has no valid range, that is, the data may increase or decrease over time, or change continuously over a large range, it is necessary to build a reference historical data set, remove singular points based on the data change trend in the recent period of time, and take the most recently collected system status data. , amount of data The following constraints must be met: ; in, represents the reference historical dataset, Indicates the first elements, the data in the reference historical data set is updated in real time as the data is collected. After the reference historical data set is determined, there is a singular point elimination method: ; in, Indicates the system status data obtained from the last sampling.

5. The method for real-time processing of collected system status data according to claim 1, characterized in that: In step 3, when constructing the smoothed attention data set, the most recently collected system status data is taken , amount of data The following constraints are met: ; in, represents a smooth attention dataset, represents the i-th element in the smoothing attention dataset.

6. The method for real-time processing of collected system status data according to claim 1, characterized in that: When constructing the integral attention data set, take the most recent collected and integrated system status data , amount of data The following constraints need to be met: ; in, represents the integral attention dataset, Indicates the first elements.

7. The method for real-time processing of collected system status data according to claim 6, characterized in that: In step 4, the steps for obtaining the curve used for fitting are:

21. Set the curve used for fitting as: ; in: It represents the first data, and is the characteristic parameter of the curve; when and When the following requirements are met, the sample points are closest to the curve used for fitting, that is: ; in: represents the characteristic parameters of the curve calculated by least squares fitting; 22. Order ,make right and Find the partial derivative and solve it jointly to get: ; ; 2.

3. The curve used for fitting is: .

8. The method for real-time processing of collected system status data according to claim 7, characterized in that: The data in the integral focus data set is updated in real time as the data is collected; the data in the smoothing focus data set is updated in real time as the data is collected.