A control feedback data verification processing method based on linear regression

Abstract

The present invention relates to the field of comprehensive analysis and verification of control feedback data, and in particular to a control feedback data verification and processing method based on linear regression, comprising: using real-time control feedback data to obtain corresponding historical abnormal control feedback data to establish a control feedback data adaptive mapping database; obtaining a data linear marking point set of the real-time control feedback data based on linear regression according to the control feedback data adaptive mapping database; performing verification processing on the data linear marking point set of the real-time control feedback data to obtain a control feedback data verification processing result, constructing a preliminary analysis model based on linear regression, deeply exploring the linear relationship between each feature and the target variable, and repeatedly testing the model through a cross-validation method to provide a basis for subsequent analysis, and on the basis of model establishment and verification, continuously and iteratively adjusting and optimizing the model parameters to improve the generalization and processing capabilities of the model.

Description

Technical Field

[0001] The present invention relates to the field of comprehensive analysis and verification of control feedback data, and in particular to a control feedback data verification processing method based on linear regression. Background Art

[0002] In most electronic medical devices, the inspection or operation process relies on signal instructions to be issued. Due to the high rigor of the steps involved, the accuracy of the operation instructions and feedback data is extremely high. At the same time, the error detection of various characteristics also requires real-time dynamic control. In addition to the problems of singleness and low accuracy of the control feedback data in the device signal in traditional verification schemes, there may also be sample bias, excessive system dependence, lack of dynamic adjustment mechanism and safety risks. Summary of the invention

[0003] In view of the deficiencies in the prior art, the present invention provides a control feedback data verification and processing method based on linear regression, which compares and analyzes data and their characteristics over a period of time through linear regression to quickly and accurately obtain verification and processing results.

[0004] To achieve the above object, the present invention provides a control feedback data verification processing method based on linear regression, comprising:

[0005] S1. Using real-time control feedback data to obtain corresponding historical abnormal control feedback data to establish a control feedback data adaptive mapping database;

[0006] S2, obtaining a data linear marking point set of real-time control feedback data based on linear regression according to the control feedback data adaptive mapping database;

[0007] S3. Perform verification processing on the data linear marking point set of the real-time control feedback data to obtain a control feedback data verification processing result.

[0008] Preferably, the method of using the real-time control feedback data to obtain the corresponding historical abnormal control feedback data to establish a control feedback data adaptability mapping database includes:

[0009] Collect real-time control feedback data;

[0010] Acquire corresponding real-time control input data according to the real-time control feedback data;

[0011] Obtaining a real-time control feedback error using the real-time control input data and the real-time control feedback data;

[0012] Acquire a corresponding signal frequency as a real-time control input label according to the real-time control input data;

[0013] Acquire a corresponding signal frequency as a real-time control feedback label according to the real-time control feedback data;

[0014] Acquiring corresponding historical abnormal control feedback data according to the real-time control feedback data;

[0015] Acquiring corresponding historical abnormal control input data according to the historical abnormal control feedback data;

[0016] Obtaining a historical abnormal control feedback error using the historical abnormal control input data and the historical abnormal control feedback data;

[0017] Acquire a corresponding signal frequency as a historical abnormal control input label according to the historical abnormal control input data;

[0018] Acquire a corresponding signal frequency as a historical abnormal control feedback label according to the historical abnormal control feedback data;

[0019] Establishing a real-time control data mapping using the real-time control feedback data and the real-time control input data;

[0020] Establishing a historical abnormal control data mapping using the historical abnormal control feedback data and the historical abnormal control input data;

[0021] Establishing a real-time input-feedback frequency mapping using the real-time control input tag and the real-time control feedback tag;

[0022] Establishing a historical abnormal input-feedback frequency mapping using the historical abnormal control input label and the historical abnormal control feedback label;

[0023] The real-time control data mapping, historical abnormal control data mapping, real-time input-feedback frequency mapping, historical abnormal input-feedback frequency mapping, real-time control feedback error and historical abnormal control feedback error are used as a control feedback data adaptive mapping database.

[0024] Furthermore, the data linear marking point set of real-time control feedback data is obtained based on linear regression according to the control feedback data adaptive mapping database, including:

[0025] S2-1, using the control feedback data adaptive mapping database to establish a continuous data analysis model for real-time control feedback data based on linear regression;

[0026] S2-2, using the control feedback data adaptive mapping database to establish a data fluctuation screening model for real-time control feedback data based on linear regression;

[0027] S2-3, using the continuous data analysis model of the real-time control feedback data and the data fluctuation screening model to perform comparison processing to obtain a data linear marking point set of the real-time control feedback data.

[0028] Furthermore, using the control feedback data adaptive mapping database to establish a continuous data analysis model for real-time control feedback data based on linear regression includes:

[0029] The real-time control data mapping of the control feedback data adaptive mapping database is divided into a first data set;

[0030] Dividing into a second data set using the real-time input-feedback frequency mapping of the control feedback data adaptive mapping database;

[0031] Using the real-time control feedback error of the control feedback data adaptive mapping database to divide into a third data set;

[0032] Using the first data set as input and the second data set corresponding to the first data set as output, training is performed based on linear regression to obtain a basic data-frequency association model of real-time control feedback data;

[0033] Using the first data set and the second data set as input, and the first data set and the second data set corresponding to a third data set as output, training based on linear regression to obtain a basic error analysis model for real-time control feedback data;

[0034] The basic data-frequency correlation model and the basic error analysis model of the real-time control feedback data are used as the continuous data analysis model of the real-time control feedback data.

[0035] Furthermore, using the control feedback data adaptive mapping database to establish a data fluctuation screening model for real-time control feedback data based on linear regression includes:

[0036] S2-2-1, using the control feedback data adaptive mapping database to map the historical abnormal control data into a fourth data set;

[0037] S2-2-2, using the historical abnormal input-feedback frequency mapping of the control feedback data adaptive mapping database to divide into a fifth data set;

[0038] S2-2-3, using the control feedback data adaptive mapping database to divide the historical abnormal control feedback errors into a sixth data set;

[0039] S2-2-4, using the fourth data set as input and the fifth data set corresponding to the fourth data set as output, training based on linear regression to obtain a primary data fluctuation screening model;

[0040] S2-2-5, using the fifth data set as input and the sixth data set corresponding to the fifth data set as output, training based on linear regression to obtain a secondary data fluctuation screening model;

[0041] S2-2-6. Determine whether the historical abnormal control data mapping corresponding to the model output of the primary data fluctuation screening model and the historical abnormal input-feedback frequency mapping corresponding to the model output of the secondary data fluctuation screening model correspond to each other. If so, use the primary data fluctuation screening model and the secondary data fluctuation screening model as data fluctuation screening models for real-time control feedback data. Otherwise, return to S2-2-4.

[0042] Furthermore, the data linear marking point set of the real-time control feedback data obtained by comparing the continuous data analysis model of the real-time control feedback data with the data fluctuation screening model includes:

[0043] S2-3-1. Obtain corresponding basic data-frequency correlation results and basic error analysis results in sequence according to the continuous data analysis model of the real-time control feedback data;

[0044] S2-3-2, sequentially obtaining corresponding primary data fluctuation screening results and secondary data fluctuation screening results according to the data fluctuation screening model of the real-time control feedback data;

[0045] S2-3-3, determine whether the basic data-frequency association result corresponds to the primary data fluctuation screening result, if so, obtain the real-time control data mapping corresponding to the primary data fluctuation screening result in the basic data-frequency association result as the data linear marking point set of the real-time control feedback data, otherwise, execute S2-3-4;

[0046] S2-3-4. Determine whether the basic error analysis result corresponds to the secondary data fluctuation screening result. If so, obtain the real-time control data mapping corresponding to the secondary data fluctuation screening result in the basic error analysis result as the data linear marking point set of the real-time control feedback data. Otherwise, the data linear marking point set of the real-time control feedback data is empty.

[0047] Further, the verification processing is performed according to the data linear marking point set of the real-time control feedback data to obtain the control feedback data verification processing result, which includes:

[0048] S3-1, using the data linear marking point set of the real-time control feedback data to perform fluctuation extraction processing to obtain an abnormal data linear marking point set of the real-time control feedback data;

[0049] S3-2, using the abnormal data linear marking point set of the real-time control feedback data to perform analysis and verification processing to obtain a control feedback data verification processing result.

[0050] Furthermore, the abnormal data linear marking point set of the real-time control feedback data is obtained by performing fluctuation extraction processing using the data linear marking point set of the real-time control feedback data, including:

[0051] S3-1-1, determine whether the data linear marking point set of the real-time control feedback data is empty, if so, execute S3-1-2, otherwise, directly execute S3-1-3;

[0052] S3-1-2, determine whether the real-time control feedback error corresponding to the real-time control feedback data is 0, if so, the abnormal data linear marking point set of the real-time control feedback data is empty, otherwise, obtain the corresponding moment of the real-time control input data when the real-time control feedback error is not 0 as the abnormal data linear marking point set of the real-time control feedback data;

[0053] S3-1-3, using the data linear marking point set of the real-time control feedback data to establish a data linear marking point function based on Matlab;

[0054] S3-1-4. Determine whether the data linear marking point function is differentiable. If so, use the data linear marking point set of the real-time control feedback data as the abnormal data linear marking point set of the real-time control feedback data. Otherwise, obtain the corresponding time of the data linear marking point set of the real-time control feedback data as the abnormal data linear marking point set of the real-time control feedback data.

[0055] Furthermore, the abnormal data linear marking point set of the real-time control feedback data is used to perform analysis and verification processing to obtain the control feedback data verification processing result, including:

[0056] S3-2-1, determine whether the abnormal data linear marking point set of the real-time control feedback data corresponds to the time, if so, execute S3-2-2, otherwise, the control feedback data verification processing result is abnormal, and output the abnormal data linear marking point set;

[0057] S3-2-2, acquiring corresponding real-time abnormal control input data and historical abnormal control input data according to the abnormal data linear marking point set;

[0058] S3-2-3, acquiring the real-time control input data of adjacent moments respectively according to the real-time abnormal control input data to obtain the real-time data trend of the control input data;

[0059] S3-2-4, obtaining historical control input data at adjacent moments according to the historical abnormal control input data to obtain historical data trends of the control input data;

[0060] S3-2-5, judging whether the real-time data trend of the control input data is consistent with the historical data trend, if so, the control feedback data verification processing result is abnormal, and outputs the real-time abnormal control input data, otherwise, the control feedback data verification processing result is normal;

[0061] The data trend refers to the change in the values ​​of adjacent data.

[0062] Compared with the closest prior art, the present invention has the following beneficial effects:

[0063] A preliminary analysis model is constructed based on linear regression, and the linear relationship between each feature and the target variable is deeply explored. The model is tested repeatedly through cross-validation to ensure its consistent performance on different data subsets, thereby verifying the stability of the model. According to the statistical results of the model output, the key features that have a significant impact on the target variable are identified and screened to provide a basis for subsequent analysis. On the basis of model establishment and verification, the model parameters are adjusted and optimized in a continuous iterative cycle to improve the generalization and processing capabilities of the model. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 It is a flow chart of a control feedback data verification processing method based on linear regression provided by the present invention. DETAILED DESCRIPTION

[0065] The specific implementation modes of the present invention will be further described in detail below in conjunction with the accompanying drawings.

[0066] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are 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.

[0067] Embodiment 1:

[0068] The present invention provides a control feedback data verification processing method based on linear regression, such as Figure 1 As shown, including:

[0069] S1. Using real-time control feedback data to obtain corresponding historical abnormal control feedback data to establish a control feedback data adaptive mapping database;

[0070] S2, obtaining a data linear marking point set of real-time control feedback data based on linear regression according to the control feedback data adaptive mapping database;

[0071] S3. Perform verification processing on the data linear marking point set of the real-time control feedback data to obtain a control feedback data verification processing result.

[0072] S1 specifically includes:

[0073] S1-1, collect real-time control feedback data;

[0074] S1-2, acquiring corresponding real-time control input data according to the real-time control feedback data;

[0075] S1-3, obtaining a real-time control feedback error by using the real-time control input data and the real-time control feedback data;

[0076] S1-4, acquiring a corresponding signal frequency as a real-time control input label according to the real-time control input data;

[0077] S1-5, acquiring a corresponding signal frequency as a real-time control feedback label according to the real-time control feedback data;

[0078] S1-6, acquiring corresponding historical abnormal control feedback data according to the real-time control feedback data;

[0079] S1-7, acquiring corresponding historical abnormal control input data according to the historical abnormal control feedback data;

[0080] S1-8, using the historical abnormal control input data and the historical abnormal control feedback data to obtain a historical abnormal control feedback error;

[0081] S1-9, acquiring a corresponding signal frequency as a historical abnormal control input label according to the historical abnormal control input data;

[0082] S1-10, acquiring a corresponding signal frequency as a historical abnormal control feedback label according to the historical abnormal control feedback data;

[0083] S1-11, establishing a real-time control data mapping using the real-time control feedback data and the real-time control input data;

[0084] S1-12, using the historical abnormal control feedback data and the historical abnormal control input data to establish a historical abnormal control data mapping;

[0085] S1-13, using the real-time control input tag and the real-time control feedback tag to establish a real-time input-feedback frequency mapping;

[0086] S1-14, using the historical abnormal control input label and the historical abnormal control feedback label to establish a historical abnormal input-feedback frequency mapping;

[0087] S1-15. Utilize the real-time control data mapping, historical abnormal control data mapping, real-time input-feedback frequency mapping, historical abnormal input-feedback frequency mapping, real-time control feedback error and historical abnormal control feedback error as a control feedback data adaptive mapping database.

[0088] S2 specifically includes:

[0089] S2-1, using the control feedback data adaptive mapping database to establish a continuous data analysis model for real-time control feedback data based on linear regression;

[0090] S2-2, using the control feedback data adaptive mapping database to establish a data fluctuation screening model for real-time control feedback data based on linear regression;

[0091] S2-3, using the continuous data analysis model of the real-time control feedback data and the data fluctuation screening model to perform comparison processing to obtain a data linear marking point set of the real-time control feedback data.

[0092] S2-1 specifically includes:

[0093] S2-1-1, using the real-time control data mapping of the control feedback data adaptive mapping database to be divided into a first data set;

[0094] S2-1-2, using the real-time input-feedback frequency mapping of the control feedback data adaptive mapping database to divide into a second data set;

[0095] S2-1-3, using the real-time control feedback error of the control feedback data adaptive mapping database to be divided into a third data set;

[0096] S2-1-4, using the first data set as input and the second data set corresponding to the first data set as output, training based on linear regression to obtain a basic data-frequency association model of real-time control feedback data;

[0097] S2-1-5, using the first data set and the second data set as input, and the first data set and the second data set corresponding to the third data set as output, training based on linear regression to obtain a basic error analysis model for real-time control feedback data;

[0098] S2-1-6. Utilize the basic data-frequency correlation model and basic error analysis model of the real-time control feedback data as a continuous data analysis model of the real-time control feedback data.

[0099] S2-2 specifically includes:

[0100] S2-2-1, using the control feedback data adaptive mapping database to map the historical abnormal control data into a fourth data set;

[0101] S2-2-2, using the historical abnormal input-feedback frequency mapping of the control feedback data adaptive mapping database to divide into a fifth data set;

[0102] S2-2-3, using the control feedback data adaptive mapping database to divide the historical abnormal control feedback errors into a sixth data set;

[0103] S2-2-4, using the fourth data set as input and the fifth data set corresponding to the fourth data set as output, training based on linear regression to obtain a primary data fluctuation screening model;

[0104] S2-2-5, using the fifth data set as input and the sixth data set corresponding to the fifth data set as output, training based on linear regression to obtain a secondary data fluctuation screening model;

[0105] S2-2-6. Determine whether the historical abnormal control data mapping corresponding to the model output of the primary data fluctuation screening model and the historical abnormal input-feedback frequency mapping corresponding to the model output of the secondary data fluctuation screening model correspond to each other. If so, use the primary data fluctuation screening model and the secondary data fluctuation screening model as data fluctuation screening models for real-time control feedback data. Otherwise, return to S2-2-4.

[0106] S2-3 specifically includes:

[0107] S2-3-1. Obtain corresponding basic data-frequency correlation results and basic error analysis results in sequence according to the continuous data analysis model of the real-time control feedback data;

[0108] S2-3-2, sequentially obtaining corresponding primary data fluctuation screening results and secondary data fluctuation screening results according to the data fluctuation screening model of the real-time control feedback data;

[0109] S2-3-3, determine whether the basic data-frequency association result corresponds to the primary data fluctuation screening result, if so, obtain the real-time control data mapping corresponding to the primary data fluctuation screening result in the basic data-frequency association result as the data linear marking point set of the real-time control feedback data, otherwise, execute S2-3-4;

[0110] S2-3-4. Determine whether the basic error analysis result corresponds to the secondary data fluctuation screening result. If so, obtain the real-time control data mapping corresponding to the secondary data fluctuation screening result in the basic error analysis result as the data linear marking point set of the real-time control feedback data. Otherwise, the data linear marking point set of the real-time control feedback data is empty.

[0111] S3 specifically includes:

[0112] S3-1, using the data linear marking point set of the real-time control feedback data to perform fluctuation extraction processing to obtain an abnormal data linear marking point set of the real-time control feedback data;

[0113] S3-2, using the abnormal data linear marking point set of the real-time control feedback data to perform analysis and verification processing to obtain a control feedback data verification processing result.

[0114] S3-1 specifically includes:

[0115] S3-1-1, determine whether the data linear marking point set of the real-time control feedback data is empty, if so, execute S3-1-2, otherwise, directly execute S3-1-3;

[0116] S3-1-2, determine whether the real-time control feedback error corresponding to the real-time control feedback data is 0, if so, the abnormal data linear marking point set of the real-time control feedback data is empty, otherwise, obtain the corresponding moment of the real-time control input data when the real-time control feedback error is not 0 as the abnormal data linear marking point set of the real-time control feedback data;

[0117] S3-1-3, using the data linear marking point set of the real-time control feedback data to establish a data linear marking point function based on Matlab;

[0118] S3-1-4. Determine whether the data linear marking point function is differentiable. If so, use the data linear marking point set of the real-time control feedback data as the abnormal data linear marking point set of the real-time control feedback data. Otherwise, obtain the corresponding time of the data linear marking point set of the real-time control feedback data as the abnormal data linear marking point set of the real-time control feedback data.

[0119] S3-2 specifically includes:

[0120] S3-2-1, determine whether the abnormal data linear marking point set of the real-time control feedback data corresponds to the time, if so, execute S3-2-2, otherwise, the control feedback data verification processing result is abnormal, and output the abnormal data linear marking point set;

[0121] S3-2-2, acquiring corresponding real-time abnormal control input data and historical abnormal control input data according to the abnormal data linear marking point set;

[0122] S3-2-3, acquiring the real-time control input data of adjacent moments respectively according to the real-time abnormal control input data to obtain the real-time data trend of the control input data;

[0123] S3-2-4, obtaining historical control input data at adjacent moments according to the historical abnormal control input data to obtain historical data trends of the control input data;

[0124] S3-2-5, judging whether the real-time data trend of the control input data is consistent with the historical data trend, if so, the control feedback data verification processing result is abnormal, and outputs the real-time abnormal control input data, otherwise, the control feedback data verification processing result is normal;

[0125] The data trend refers to the change in the values ​​of adjacent data.

[0126] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0127] The present invention is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0128] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0129] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0130] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A control feedback data verification processing method based on linear regression, characterized in that: include: S1. Using real-time control feedback data to obtain corresponding historical abnormal control feedback data to establish a control feedback data adaptive mapping database; S1-1, collect real-time control feedback data; S1-2, acquiring corresponding real-time control input data according to the real-time control feedback data; S1-3, obtaining a real-time control feedback error by using the real-time control input data and the real-time control feedback data; S1-4, acquiring a corresponding signal frequency as a real-time control input label according to the real-time control input data; S1-5, acquiring a corresponding signal frequency as a real-time control feedback label according to the real-time control feedback data; S1-6, acquiring corresponding historical abnormal control feedback data according to the real-time control feedback data; S1-7, acquiring corresponding historical abnormal control input data according to the historical abnormal control feedback data; S1-8, using the historical abnormal control input data and the historical abnormal control feedback data to obtain a historical abnormal control feedback error; S1-9, acquiring a corresponding signal frequency as a historical abnormal control input label according to the historical abnormal control input data; S1-10, acquiring a corresponding signal frequency as a historical abnormal control feedback label according to the historical abnormal control feedback data; S1-11, establishing a real-time control data mapping using the real-time control feedback data and the real-time control input data; S1-12, using the historical abnormal control feedback data and the historical abnormal control input data to establish a historical abnormal control data mapping; S1-13, using the real-time control input tag and the real-time control feedback tag to establish a real-time input-feedback frequency mapping; S1-14, using the historical abnormal control input label and the historical abnormal control feedback label to establish a historical abnormal input-feedback frequency mapping; S1-15, using the real-time control data mapping, historical abnormal control data mapping, real-time input-feedback frequency mapping, historical abnormal input-feedback frequency mapping, real-time control feedback error and historical abnormal control feedback error as a control feedback data adaptive mapping database; S2, obtaining a data linear marking point set of real-time control feedback data based on linear regression according to the control feedback data adaptive mapping database; S2-1, using the control feedback data adaptive mapping database to establish a continuous data analysis model for real-time control feedback data based on linear regression; S2-1-1, using the real-time control data mapping of the control feedback data adaptive mapping database to be divided into a first data set; S2-1-2, using the real-time input-feedback frequency mapping of the control feedback data adaptive mapping database to divide into a second data set; S2-1-3, using the real-time control feedback error of the control feedback data adaptive mapping database to be divided into a third data set; S2-1-4, using the first data set as input and the second data set corresponding to the first data set as output, training based on linear regression to obtain a basic data-frequency association model of real-time control feedback data; S2-1-5, using the first data set and the second data set as input, and the first data set and the second data set corresponding to the third data set as output, training based on linear regression to obtain a basic error analysis model for real-time control feedback data; S2-1-6, using the basic data-frequency correlation model and the basic error analysis model of the real-time control feedback data as a continuous data analysis model of the real-time control feedback data; S2-2, using the control feedback data adaptive mapping database to establish a data fluctuation screening model for real-time control feedback data based on linear regression; S2-2-1, using the control feedback data adaptive mapping database to map the historical abnormal control data into a fourth data set; S2-2-2, using the historical abnormal input-feedback frequency mapping of the control feedback data adaptive mapping database to divide into a fifth data set; S2-2-3, using the control feedback data adaptive mapping database to divide the historical abnormal control feedback errors into a sixth data set; S2-2-4, using the fourth data set as input and the fifth data set corresponding to the fourth data set as output, training based on linear regression to obtain a primary data fluctuation screening model; S2-2-5, using the fifth data set as input and the sixth data set corresponding to the fifth data set as output, training based on linear regression to obtain a secondary data fluctuation screening model; S2-2-6, determine whether the historical abnormal control data mapping corresponding to the model output of the primary data fluctuation screening model and the historical abnormal input-feedback frequency mapping corresponding to the model output of the secondary data fluctuation screening model correspond to each other, if so, use the primary data fluctuation screening model and the secondary data fluctuation screening model as the data fluctuation screening model for real-time control feedback data, otherwise, return to S2-2-4; S2-3, using the continuous data analysis model of the real-time control feedback data and the data fluctuation screening model to perform comparison processing to obtain a data linear marking point set of the real-time control feedback data; S2-3-1. Obtain corresponding basic data-frequency correlation results and basic error analysis results in sequence according to the continuous data analysis model of the real-time control feedback data; S2-3-2, sequentially obtaining corresponding primary data fluctuation screening results and secondary data fluctuation screening results according to the data fluctuation screening model of the real-time control feedback data; S2-3-3, determine whether the basic data-frequency association result corresponds to the primary data fluctuation screening result, if so, obtain the real-time control data mapping corresponding to the primary data fluctuation screening result in the basic data-frequency association result as the data linear marking point set of the real-time control feedback data, otherwise, execute S2-3-4; S2-3-4, determine whether the basic error analysis result corresponds to the secondary data fluctuation screening result, if so, obtain the real-time control data mapping corresponding to the secondary data fluctuation screening result in the basic error analysis result as the data linear marking point set of the real-time control feedback data, otherwise, the data linear marking point set of the real-time control feedback data is empty; S3, performing verification processing according to the data linear marking point set of the real-time control feedback data to obtain a control feedback data verification processing result; S3-1, using the data linear marking point set of the real-time control feedback data to perform fluctuation extraction processing to obtain an abnormal data linear marking point set of the real-time control feedback data; S3-2, using the abnormal data linear marking point set of the real-time control feedback data to perform analysis and verification processing to obtain a control feedback data verification processing result.

2. A control feedback data verification processing method based on linear regression as claimed in claim 1, characterized in that: The abnormal data linear marking point set of the real-time control feedback data is obtained by performing fluctuation extraction processing using the data linear marking point set of the real-time control feedback data, including: S3-1-1, determine whether the data linear marking point set of the real-time control feedback data is empty, if so, execute S3-1-2, otherwise, directly execute S3-1-3; S3-1-2, determine whether the real-time control feedback error corresponding to the real-time control feedback data is 0, if so, the abnormal data linear marking point set of the real-time control feedback data is empty, otherwise, obtain the corresponding moment of the real-time control input data when the real-time control feedback error is not 0 as the abnormal data linear marking point set of the real-time control feedback data; S3-1-3, using the data linear marking point set of the real-time control feedback data to establish a data linear marking point function based on Matlab; S3-1-4. Determine whether the data linear marking point function is differentiable. If so, use the data linear marking point set of the real-time control feedback data as the abnormal data linear marking point set of the real-time control feedback data. Otherwise, obtain the corresponding time of the data linear marking point set of the real-time control feedback data as the abnormal data linear marking point set of the real-time control feedback data.

3. A control feedback data verification processing method based on linear regression as claimed in claim 2, characterized in that: The control feedback data verification processing result obtained by analyzing and verifying the abnormal data linear marking point set of the real-time control feedback data includes: S3-2-1, determine whether the abnormal data linear marking point set of the real-time control feedback data corresponds to the time, if so, execute S3-2-2, otherwise, the control feedback data verification processing result is abnormal, and output the abnormal data linear marking point set; S3-2-2, acquiring corresponding real-time abnormal control input data and historical abnormal control input data according to the abnormal data linear marking point set; S3-2-3, acquiring the real-time control input data of adjacent moments respectively according to the real-time abnormal control input data to obtain the real-time data trend of the control input data; S3-2-4, obtaining historical control input data at adjacent moments according to the historical abnormal control input data to obtain historical data trends of the control input data; S3-2-5, judging whether the real-time data trend of the control input data is consistent with the historical data trend, if so, the control feedback data verification processing result is abnormal, and outputs the real-time abnormal control input data, otherwise, the control feedback data verification processing result is normal; The data trend refers to the change in the values ​​of adjacent data.

Images