Relay protection equipment constant value calibration system and method thereof

Abstract

The invention provides a relay protection equipment constant value calibration system and method, and relates to the technical field of constant value calibration, and the system comprises a data collection module, a data processing module, a constant value calibration model construction module, a constant value calibration module, and a constant value calibration model adjustment module. The constant value calibration model adjusting module is electrically connected with the constant value calibration model building module and is used for collecting a calibration accuracy score omega 1 of a user and an equipment operation effect score omega 2, generating a user satisfaction coefficient omega, adjusting parameters and an objective function of a constant value calibration model according to the user satisfaction coefficient omega, and setting the constant value calibration model according to the parameters and the objective function. According to the method, the reference factor of user feedback is introduced, so that the constant value can be automatically optimized according to the historical data of the equipment, the real-time monitoring result and the user feedback, and the calculated constant value data can better meet the requirements of the user.

Description

Technical Field

[0001] The present invention relates to the technical field of fixed value calibration, and in particular to a fixed value calibration system and method for relay protection equipment. Background Art

[0002] Relay protection plays an increasingly important role in the safe and stable operation of power grids. The reliable operation and maintenance management of relay protection equipment is an important support for power network infrastructure and power safety assurance. The large number of relay protection equipment, the complex types, the strict logic, and the massive amount of information all indicate that the professional management of relay protection is a complex and huge system engineering, and the demand for its refined management is becoming more and more urgent. However, at present, the verification of relay protection equipment values ​​on site mostly adopts paper value sheets and manual verification, which has the problems of large workload and easy errors.

[0003] In the prior art, there is a system and method for intelligent calibration of fixed value information of relay protection equipment with publication number "CN112003234A" (classification number H02H3 / 00), which includes: obtaining a parameter list and forming image data; forming a parameter comparison table based on the image data content; obtaining the fixed value information in the relay protection equipment and displaying it in a table form; performing a data comparison of the same type based on the fixed value information and the content of the parameter comparison table, and forming a conclusion report. By using a mobile terminal as an intermediate medium in an offline state, the parameter list and the fixed value values ​​in the relay protection equipment are calibrated in the terminal, and displayed in the form of a report, thereby realizing intelligent calibration of the values ​​in the equipment in an offline state, avoiding the problems of large workload and easy errors in manual verification.

[0004] However, the existing technology still has major defects. For example, the fixed value data used as a calibration reference in the existing technology does not introduce the reference factor of user satisfaction feedback, and the selection and calculation of the fixed value data cannot be adjusted according to user feedback. There are problems with the fixed value data used as a calibration reference, which has an adverse impact on the subsequent fixed value calibration work. Summary of the invention

[0005] The object of the present invention is to provide a relay protection device setting calibration system and method thereof to solve the problems raised in the above background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A relay protection device setting calibration system, comprising:

[0008] A data acquisition module, wherein the data acquisition unit is used to collect the operation data of the equipment and the setting data of the relay protection device in real time;

[0009] A data processing module, which is electrically connected to the data acquisition module and is used to process the operating data and fixed value data transmitted by the data acquisition module, and divide the processed results into a training set, a test set and a calibration set;

[0010] A fixed value calibration model construction module, the fixed value calibration model construction module is electrically connected to the data processing module, and is used to train a fixed value calibration model according to a training set, and to evaluate and test the trained fixed value calibration model according to a test set;

[0011] A fixed value calibration module, the fixed value calibration module is electrically connected to the data processing module, the fixed value calibration model construction module, and the relay protection device, and is used to store the fixed value calibration model constructed by the fixed value calibration model construction unit, process the calibration centralized data to obtain adjustment data, and transmit the adjustment data to the relay protection device for fixed value calibration;

[0012] The fixed value calibration model adjustment module is electrically connected to the fixed value calibration model construction module, and is used to collect the user's calibration accuracy score ω1 and the equipment operation effect score ω2, generate the user satisfaction coefficient ω, and adjust the parameters and objective function of the fixed value calibration model according to the user satisfaction coefficient ω.

[0013] Preferably, the operating data includes current I t , voltage U t , and frequency f t The fixed value data includes the current fixed value Voltage setting And the frequency setting f t d , where t represents the number of different moments, t = 1, 2, 3, ..., n.

[0014] Preferably, the data processing module includes a data merging unit, a data standardization unit, a feature engineering unit, and a data set division unit:

[0015] A data merging unit, which is electrically connected to the data acquisition module, is used to merge the operating data except for the nth moment and the fixed value data of the relay protection device into the historical data set matrix X, where X=(x 1 , x 2 , …, x t , …, x n-1 ), and merge the operation data of the equipment at time n into the real-time operation data set z1 n The setting data of the relay protection device at time n is merged into the real-time setting data set z2 n middle;

[0016] A data standardization unit, the data standardization unit is electrically connected to the data merging unit, and is used to normalize the real-time running data set z1 n , data set x in historical data set matrix X t The data is standardized;

[0017] Among them, for the data set x t The calculation formula for standardizing the data in is as follows:

[0018]

[0019]

[0020]

[0021] It should be noted that For n-1 data sets x t The mean value of x s For n-1 data sets x t The standard deviation of x t ' is x t The dataset has been standardized;

[0022] Among them, for the real-time running data set z1 n The calculation formula for standardizing the data in is as follows:

[0023]

[0024] It should be noted that z1 n ' is z1 n The dataset has been standardized;

[0025] The feature engineering unit is electrically connected to the data acquisition module and is used to extract the phase angle difference between the current and the voltage, which can be processed by the following formula:

[0026] XOtM t =arg(U t )-arg(I t )

[0027] It should be noted that arg(U t ) represents the phase of the voltage at time t, arg(I t ) represents the phase of the current at time t, XWJ t It represents the phase angle difference between current and voltage at time t;

[0028] The data set division unit is electrically connected to the data standardization unit and the feature engineering unit, and is used to divide the standardized data set x t', and the phase angle difference XWJ of the current and voltage except time n t The merged data is divided into training and test sets and used to transform z1 n ', XWJ n , and z2 n Merge into calibration set;

[0029] The training set includes X sets of data, and the test set includes Y sets of data. One set of data is the standardized data set x at the same time. t ', and the phase angle difference between current and voltage XWJ t .

[0030] Preferably, the fixed value calibration module includes a calibration set input unit, an adjustment data calculation unit, and an adjustment data output unit;

[0031] A calibration set input unit, the calibration set input unit is electrically connected to the data processing module and the adjustment data calculation unit, and is used to obtain the calibration set in the data processing module and input the calibration set into the adjustment data calculation unit;

[0032] The adjustment data calculation unit is electrically connected to the fixed value calibration model construction module and the calibration set input unit, and is used to store the fixed value calibration model constructed by the fixed value calibration model construction module, and adjust the z1 in the calibration set through the objective function F of the fixed value calibration model. n ', XWJ n Perform calculations to obtain the predicted constant value z2 at time n n ', and then predict the constant value z2 n ' and z2 n Get adjustment dataσ n , adjust the data σ n The calculation formula is as follows:

[0033] σ n =z2 n '-z2 n

[0034] The adjustment data output unit is electrically connected to the adjustment data calculation unit and the relay protection device, and is used to obtain the adjustment data σ calculated by the adjustment data calculation unit. n , adjust the data σ n Input into the relay protection device to calibrate the set value data in the relay protection device.

[0035] Preferably, the fixed value calibration model adjustment module includes a user feedback information collection unit, a user satisfaction calculation unit, a fixed value calibration model first adjustment unit, and a fixed value calibration model second adjustment unit.

[0036] Preferably, the user feedback information collection unit is used to collect the user's calibration accuracy score ω1 and the device operation effect score ω2, as follows:

[0037] Establish a user feedback system that includes calibration accuracy scoring and equipment operation effect scoring. The rules for calibration accuracy scoring and equipment operation effect scoring are both 1-9 points, with 1 point representing the best and 9 points representing the worst.

[0038] Preferably, the user satisfaction calculation unit is electrically connected to the user feedback information collection unit, and is used to generate a user satisfaction coefficient ω according to the calibration accuracy score ω1 and the equipment operation effect score ω2, as follows:

[0039] Based on ω1+ω2=1, ω1 and ω2 are scaled and standardized to obtain ω1' and ω2', and the user satisfaction coefficient ω is generated. The calculation formula of the user satisfaction coefficient ω is as follows:

[0040] ω=α1*ω1'+α2*ω2'

[0041] It should be noted that α1 and α2 are preset proportional coefficients of ω1′ and ω2′, and 0<α1≤α2<1, and α1+α2=1.

[0042] Preferably, the first adjustment unit of the fixed value calibration model is electrically connected to the user satisfaction calculation unit and the fixed value calibration model construction module, and is used to obtain the user satisfaction coefficient ω calculated by the user satisfaction calculation unit, and transmit the user satisfaction coefficient ω to the fixed value calibration model construction module, and adjust the model parameters of the fixed value calibration model. The specific adjustment formula is as follows:

[0043] λ new =λ old -σ*θ-ω*τ

[0044] It should be noted that λ old represents the old parameters of the fixed-value calibration model, λ new represents the new parameters of the fixed-value calibration model, σ represents the learning rate, which is used to control the step size of parameter update, θ represents the gradient of the objective function to the model parameters, and τ represents the gradient of the user feedback to the model parameters.

[0045] Preferably, the second adjustment unit of the fixed value calibration model is electrically connected to the user satisfaction calculation unit and the fixed value calibration model construction module, and is used to obtain the user satisfaction coefficient ω calculated by the user satisfaction calculation unit, and compare the user satisfaction coefficient ω with a preset threshold value TTH. When the user satisfaction coefficient ω is greater than the preset threshold value TTH, the user satisfaction coefficients x in the training set and the test set are eliminated. n-m ', x n-m+1 ', ..., xn-1 'data set, use other data sets that satisfy the user to calculate the objective function of the fixed value calibration model;

[0046] It should be noted that m is a natural number greater than 1.

[0047] A relay protection device fixed value calibration method, used in the above relay protection device fixed value calibration system, comprises the following steps:

[0048] S1, real-time collection of equipment operation data and relay protection device setting data;

[0049] S2, processing the operating data and the fixed value data, and dividing the processed results into a training set, a test set and a calibration set;

[0050] S3, training a fixed-value calibration model according to the training set, and performing an evaluation test on the trained fixed-value calibration model according to the test set;

[0051] S4, processing the calibration centralized data according to the fixed value calibration model to obtain adjustment data, and transmitting the adjustment data to the relay protection device for fixed value calibration;

[0052] S5, collect the user's calibration accuracy score ω1 and the equipment operation effect score ω2, generate the user satisfaction coefficient ω, and adjust the parameters and objective function of the fixed value calibration model according to the user satisfaction coefficient ω.

[0053] Compared with the prior art, the present invention has the following beneficial effects:

[0054] The relay protection device constant value calibration system and method of the present invention generates a user satisfaction coefficient based on the user's calibration accuracy score and the equipment operation effect score, and adjusts the parameters and objective function of the constant value calibration model based on the user satisfaction coefficient, thereby introducing the user feedback as a reference factor. In this way, the constant value can be automatically optimized based on the equipment's historical data, real-time monitoring results and user feedback, so that the calculated constant value data can better meet the user's needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 This is a module diagram of the relay protection device setting calibration system of the present invention;

[0056] Figure 2 This is a unit diagram of a data processing module in the present invention;

[0057] Figure 3 This is a unit diagram of a fixed value calibration module in the present invention;

[0058] Figure 4 It is a unit diagram of the fixed value calibration model adjustment module in the present invention;

[0059] Figure 5 The present invention is a flow chart of the method for calibrating the fixed value of relay protection equipment. DETAILED DESCRIPTION

[0060] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with specific embodiments.

[0061] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the present invention should be understood by people with ordinary skills in the field to which the present invention belongs. The words "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0062] Embodiment 1:

[0063] See also Figure 1-4 The present invention provides a relay protection device setting calibration system according to an embodiment, comprising:

[0064] Data acquisition module: The data acquisition unit is used to collect the operation data of the equipment and the setting data of the relay protection device in real time, as follows:

[0065] The data acquisition module collects the operation data of the equipment in real time through sensors and equipment interfaces. The operation data includes the current I t , voltage U t , and frequency f t And other data;

[0066] The data acquisition module obtains the fixed value data of the relay protection device in real time through the communication protocol. The fixed value data includes the current fixed value. Voltage setting And the frequency setting f t d and other data;

[0067] It should be noted that t represents the number of different moments, t=1, 2, 3, ..., n, n is the latest collection moment, and the intervals between different moments can be 12 hours, 1 day, 3 days, 7 days, etc., which are set according to the actual situation on site and will not be elaborated here.

[0068] The data processing module is electrically connected to the data acquisition module, and processes the operating data and fixed value data transmitted by the data acquisition module, and divides the processed results into a training set, a test set and a calibration set. The data processing module includes a data merging unit, a data standardization unit, a feature engineering unit, and a data set division unit:

[0069] The data merging unit is electrically connected to the data acquisition module and is used to merge the operation data (current I t , voltage U t , and frequency f t ) and the setting data of the relay protection device (current setting Voltage setting And the frequency setting f t d) Merge into the historical data set matrix X, where X = (x 1 , x 2 , …, x t , …, x n-1 ), and the operating data of the device at time n (current I n , voltage U n , and frequency f n ) is merged into the real-time running data set z1 n and the setting data of the relay protection device at time n (current setting value Voltage setting And frequency setting ) merged into the real-time constant value data set z2 n middle;

[0070] It should be noted that the real-time running data set z1 n Indicates the latest operating data, real-time constant value data set z2 n Indicates the latest fixed value data. As the data acquisition module continuously collects operating data and fixed value information, the real-time operating data set z1 n and real-time constant value dataset z2 n The data is constantly updated.

[0071] The data standardization unit is electrically connected to the data merging unit and is used to perform real-time operation on the data set z1. n , data set x in historical data set matrix X t Standardize the data to ensure that different data have consistent scales;

[0072] Among them, for the data set x t The data in the standardization process can be processed by using existing technology, such as the following formula:

[0073]

[0074]

[0075]

[0076] It should be noted that For n-1 data sets x t The mean value of x s For n-1 data sets x t The standard deviation of x t ' is x t The dataset has been standardized;

[0077] Among them, for the real-time running data set z1 n The calculation formula for standardizing the data in is as follows:

[0078]

[0079] It should be noted that z1 n ' is z1 n The dataset after normalization.

[0080] The feature engineering unit is electrically connected to the data acquisition module and is used to extract the phase angle difference between the current and the voltage. The phase angle difference between the current and the voltage is an important related feature of the fixed value calibration. The acquisition method can adopt the existing technology, such as processing by the following formula:

[0081] XOtM t =arg(U t )-arg(I t )

[0082] It should be noted that arg(U t ) represents the phase of the voltage at time t, arg(I t ) represents the phase of the current at time t, XWJ t Represents the phase angle difference between current and voltage at time t, arg(I t ) and arg(U t ) can be obtained through electrical testing equipment.

[0083] The data set division unit is electrically connected to the data standardization unit and the feature engineering unit, and is used to divide the standardized data set x t ', and the phase angle difference XWJ of the current and voltage except time n t The merged data is divided into training and test sets and used to transform z1 n ', XWJ n , and z2n Merge into calibration set;

[0084] The training set includes X sets of data, and the test set includes Y sets of data. One set of data is the standardized data set x at the same time. t ', and the phase angle difference between current and voltage XWJ t , X and Y can choose a ratio of 80 / 20 or 70 / 30.

[0085] The fixed value calibration model construction module is electrically connected to the data processing module and is used to train a fixed value calibration model according to the training set, and to evaluate and test the trained fixed value calibration model according to the test set, as follows:

[0086] Obtain the objective function, input the data in the training set into the machine learning model, use the linear regression method to train the machine learning model, and obtain the fixed value calibration model and its objective function F. The calculation formula of the objective function F is as follows:

[0087]

[0088] It should be noted that i represents the number of different groups of data in the training set, i = 1, 2, 3, ..., X, represents the predicted fixed value generated by the fixed value calibration model under the data calculation of the i-th training set, y i Representation and The actual setting value of the relay protection device at the same moment;

[0089] Adjust the fixed value calibration model parameters and use cross-validation and other methods to adjust the model's hyperparameters to improve the model's generalization ability. The processing formula is as follows:

[0090] λ new =λ old -σ*θ

[0091] It should be noted that λ old represents the old parameters of the fixed-value calibration model, λ new represents the new parameters of the fixed value calibration model, σ represents the learning rate, which is used to control the step size of parameter update, and θ represents the gradient of the objective function F to the model parameters. Both the learning rate σ and the gradient θ can be set according to the actual situation and will not be described here. The hyperparameters of the fixed value calibration model are continuously updated through the above method to optimize the fixed value calibration model;

[0092] The fixed value calibration model is evaluated by using the data in the test set to test the fixed value calibration model. The performance of the fixed value calibration model is evaluated through the test results to ensure the accuracy of the fixed value calibration model. Specifically, the performance of the fixed value calibration model can be evaluated through indicators such as the root mean square error and mean absolute error of the test results. Evaluation of model performance is an existing technology and will not be elaborated here.

[0093] A fixed value calibration module, the fixed value calibration module is electrically connected to the data processing module, the fixed value calibration model construction module, and the relay protection device, and is used to store the fixed value calibration model constructed by the fixed value calibration model construction unit, process the calibration set data to obtain adjustment data, and transmit the adjustment data to the relay protection device for fixed value calibration. The fixed value calibration module includes a calibration set input unit, an adjustment data calculation unit, and an adjustment data output unit;

[0094] A calibration set input unit, the calibration set input unit is electrically connected to the data processing module and the adjustment data calculation unit, and is used to obtain the calibration set in the data processing module and input the calibration set into the adjustment data calculation unit;

[0095] The adjustment data calculation unit is electrically connected to the fixed value calibration model construction module and the calibration set input unit, and is used to store the fixed value calibration model constructed by the fixed value calibration model construction module, and to adjust the z1 in the calibration set through the objective function F of the fixed value calibration model. n ', XWJ n Perform calculations to obtain the predicted constant value z2 at time n n ', and then predict the constant value z2 n ' and z2 n Get adjustment dataσ n , adjust the data σ n The calculation formula is as follows:

[0096] σ n =z2 n '-z2 n

[0097] The adjustment data output unit is electrically connected to the adjustment data calculation unit and the relay protection device, and is used to obtain the adjustment data σ calculated by the adjustment data calculation unit. n , adjust the data σ n Input into the relay protection device to calibrate the set value data in the relay protection device.

[0098] The fixed value calibration model adjustment module is electrically connected to the fixed value calibration model construction module, and is used to collect the user's calibration accuracy score ω1 and the equipment operation effect score ω2, generate the user satisfaction coefficient ω, and adjust the parameters and objective function of the fixed value calibration model according to the user satisfaction coefficient ω.

[0099] The user feedback information collection unit is used to collect the user's calibration accuracy score ω1 and the device operation effect score ω2, as follows:

[0100] Establish a user feedback system that includes calibration accuracy scoring and equipment operation effect scoring. The calibration accuracy scoring and equipment operation effect scoring rules are both 1-9 points, with 1 point representing the best and 9 points representing the worst. By collecting the calibration accuracy score ω1 and equipment operation effect score ω2 filled in by customers, the actual application of the fixed value calibration model can be judged, which is convenient for subsequent improvement of the fixed value calibration model based on user feedback;

[0101] As an implementation method, the scoring criteria are as follows

[0102] Points Scoring Criteria 1 Excellent calibration accuracy 3 Good calibration accuracy 5 Average calibration accuracy 7 Calibration accuracy error is too large 9 Calibration accuracy error is too large

[0103] Points Scoring Criteria 1 Excellent equipment operation 3 The equipment is running well 5 Equipment operation is normal 7 Poor equipment operation 9 Equipment is operating poorly

[0104] The user satisfaction calculation unit is electrically connected to the user feedback information collection unit and is used to generate a user satisfaction coefficient ω according to the calibration accuracy score ω1 and the equipment operation effect score ω2, as follows:

[0105] Based on ω1+ω2=1, ω1 and ω2 are scaled and standardized to obtain ω1' and ω2', and the user satisfaction coefficient ω is generated. The calculation formula of the user satisfaction coefficient ω is as follows

[0106] ω=α1*ω1'+α2*ω2'

[0107] It should be noted that α1 and α2 are preset proportional coefficients of ω1' and ω2', and 0<α1≤α2<1, and α1+α2=1. The specific values ​​of α1 and α2 are generally determined by technicians in this field according to actual conditions. For example, in user feedback, if the calibration accuracy score is large, α1 is adjusted upward. Similarly, if the equipment operation effect score is large, α2 is adjusted upward.

[0108] It should be noted that the larger the calibration accuracy score ω1 and the equipment operation effect score ω2 are, the less satisfied the customer is with the prediction results of the fixed value calibration model, and the larger the user satisfaction coefficient ω is.

[0109] The first adjustment unit of the fixed value calibration model is electrically connected to the user satisfaction calculation unit and the fixed value calibration model construction module, and is used to obtain the user satisfaction coefficient ω calculated by the user satisfaction calculation unit, and transmit the user satisfaction coefficient ω to the fixed value calibration model construction module, and adjust the model parameters of the fixed value calibration model. The specific adjustment formula is as follows:

[0110] λ new =λ old -σ*θ-ω*τ

[0111] It should be noted that λ old represents the old parameters of the fixed-value calibration model, λ new represents the new parameters of the fixed value calibration model, σ represents the learning rate, which is used to control the step size of parameter update, θ represents the gradient of the objective function F to the model parameters, and τ represents the gradient of the user feedback to the model parameters;

[0112] It should be noted that the smaller ω is, the more the model pays attention to historical data during the update process, and is relatively unaffected by user feedback, which is more suitable for fixed-value calibration models with higher user satisfaction. Similarly, the larger ω is, the more easily the model is affected by user feedback, the more flexible it is, and it can adapt to new changes more quickly in a short period of time, which is more suitable for fixed-value calibration models with poor user satisfaction.

[0113] The second adjustment unit of the fixed value calibration model is electrically connected to the user satisfaction calculation unit and the fixed value calibration model construction module, and is used to obtain the user satisfaction coefficient ω calculated by the user satisfaction calculation unit, and compare the user satisfaction coefficient ω with the preset threshold value TTH. When the user satisfaction coefficient ω is greater than the preset threshold value TTH, it means that the prediction effect of the fixed value calibration model at this time is not satisfactory to the user, that is, it means that the recent training set and test set have an effect that is not conducive to customer satisfaction in the fixed value calibration model, and then the x in the training set and the test set are removed. n-m ', x n-m+1 ', ..., x n-1 'data set, use other data sets that satisfy the user to calculate the objective function of the fixed value calibration model, so that the fixed value calibration model can adapt to the user's needs, thereby improving user satisfaction;

[0114] It should be noted that m is a natural number greater than 1, and the value of m can be 3, 5, 7, etc. The specific selection is based on the actual situation on site and will not be elaborated here.

[0115] Embodiment 2:

[0116] See also Figure 5 The present invention provides a method for calibrating a relay protection device setting value according to an embodiment of the present invention, comprising the following steps:

[0117] S1, real-time collection of equipment operation data and relay protection device setting data;

[0118] S2, processing the operating data and the fixed value data, and dividing the processed results into a training set, a test set and a calibration set;

[0119] S3, training a fixed-value calibration model according to the training set, and performing an evaluation test on the trained fixed-value calibration model according to the test set;

[0120] S4, processing the calibration centralized data according to the fixed value calibration model to obtain adjustment data, and transmitting the adjustment data to the relay protection device for fixed value calibration;

[0121] S5, collect the user's calibration accuracy score ω1 and the equipment operation effect score ω2, generate the user satisfaction coefficient ω, and adjust the parameters and objective function of the fixed value calibration model according to the user satisfaction coefficient ω.

[0122] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0123] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented using software, the above embodiments may be implemented in whole or in part in the form of a computer program product. Those skilled in the art may appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein may be implemented by computer software, electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed by hardware or software depends on the specific application and design constraints of the technical solution.

[0124] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, and may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0125] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

Claims

1. A relay protection device setting calibration system, characterized in that: include: A data acquisition module, wherein the data acquisition unit is used to collect the operation data of the equipment and the setting data of the relay protection device in real time; A data processing module, which is electrically connected to the data acquisition module and is used to process the operating data and fixed value data transmitted by the data acquisition module, and divide the processed results into a training set, a test set and a calibration set; A fixed value calibration model construction module, the fixed value calibration model construction module is electrically connected to the data processing module, and is used to train a fixed value calibration model according to a training set, and to evaluate and test the trained fixed value calibration model according to a test set; A fixed value calibration module, the fixed value calibration module is electrically connected to the data processing module, the fixed value calibration model construction module, and the relay protection device, and is used to store the fixed value calibration model constructed by the fixed value calibration model construction unit, process the calibration centralized data to obtain adjustment data, and transmit the adjustment data to the relay protection device for fixed value calibration; The fixed value calibration model adjustment module is electrically connected to the fixed value calibration model construction module, and is used to collect the user's calibration accuracy score ω1 and the equipment operation effect score ω2, generate the user satisfaction coefficient ω, and adjust the parameters and objective function of the fixed value calibration model according to the user satisfaction coefficient ω.

2. The relay protection device setting calibration system according to claim 1, characterized in that: The operating data includes the current I t , voltage U t , and frequency f t The fixed value data includes the current fixed value Voltage setting And the frequency setting f t d , where t represents the number of different moments, t = 1, 2, 3, ..., n.

3. The relay protection device setting calibration system according to claim 2, characterized in that: The data processing module includes a data merging unit, a data standardization unit, a feature engineering unit, and a data set division unit: A data merging unit, which is electrically connected to the data acquisition module, is used to merge the operating data except for the nth moment and the fixed value data of the relay protection device into the historical data set matrix X, where X = (x1, x2, ..., x t , …, x n-1 ), and merge the operation data of the equipment at time n into the real-time operation data set z1 n The setting data of the relay protection device at time n is merged into the real-time setting data set z2 n middle; A data standardization unit, the data standardization unit is electrically connected to the data merging unit, and is used to normalize the real-time running data set z1 n , data set x in historical data set matrix X t The data is standardized; Among them, for the data set x t The calculation formula for standardizing the data in is as follows: It should be noted that For n-1 data sets x t The mean value of x s For n-1 data sets x t The standard deviation of x t ' is x t The dataset has been standardized; Among them, for the real-time running data set z1 n The calculation formula for standardizing the data in is as follows: It should be noted that z1 n ' is z1 n The dataset has been standardized; The feature engineering unit is electrically connected to the data acquisition module and is used to extract the phase angle difference between the current and the voltage, which can be processed by the following formula: XWJ t =arg(U t )-arg(I t ) It should be noted that arg(U t ) represents the phase of the voltage at time t, arg(I t ) represents the phase of the current at time t, XWJ t It represents the phase angle difference between current and voltage at time t; The data set division unit is electrically connected to the data standardization unit and the feature engineering unit, and is used to divide the data set x after the standardization process into t ', and the phase angle difference XWJ of the current and voltage except time n t The merged data is divided into training and test sets and used to transform z1 n ', XWJ n , and z2 n Merge into calibration set; The training set includes X sets of data, and the test set includes Y sets of data. One set of data is the standardized data set x at the same time. t ', and the phase angle difference between current and voltage XWJ t .

4. The relay protection device setting calibration system according to claim 3 is characterized in that: The fixed value calibration module includes a calibration set input unit, an adjustment data calculation unit, and an adjustment data output unit; A calibration set input unit, the calibration set input unit is electrically connected to the data processing module and the adjustment data calculation unit, and is used to obtain the calibration set in the data processing module and input the calibration set into the adjustment data calculation unit; The adjustment data calculation unit is electrically connected to the fixed value calibration model construction module and the calibration set input unit, and is used to store the fixed value calibration model constructed by the fixed value calibration model construction module, and adjust the z1 in the calibration set through the objective function F of the fixed value calibration model. n ', XWJ n Perform calculations to obtain the predicted constant value z2 at time n n ', and then predict the constant value z2 n ' and z2 n Get adjustment dataσ n , adjust the data σ n The calculation formula is as follows: s n =z2 n '-z2 n The adjustment data output unit is electrically connected to the adjustment data calculation unit and the relay protection device, and is used to obtain the adjustment data σ calculated by the adjustment data calculation unit. n , adjust the data σ n Input into the relay protection device to calibrate the set value data in the relay protection device.

5. The relay protection device setting calibration system according to claim 3, characterized in that: The fixed value calibration model adjustment module includes a user feedback information collection unit, a user satisfaction calculation unit, a fixed value calibration model first adjustment unit, and a fixed value calibration model second adjustment unit.

6. The relay protection device setting calibration system according to claim 5, characterized in that: The user feedback information collection unit is used to collect the user's calibration accuracy score ω1 and the device operation effect score ω2, as follows: Establish a user feedback system that includes calibration accuracy scoring and equipment operation effect scoring. The rules for calibration accuracy scoring and equipment operation effect scoring are both 1-9 points, with 1 point representing the best and 9 points representing the worst.

7. The relay protection device setting calibration system according to claim 5, characterized in that: The user satisfaction calculation unit is electrically connected to the user feedback information collection unit, and is used to generate a user satisfaction coefficient ω according to the calibration accuracy score ω1 and the equipment operation effect score ω2, as follows: Based on ω1+ω2=1, ω1 and ω2 are scaled and standardized to obtain ω1' and ω2', and the user satisfaction coefficient ω is generated. The calculation formula of the user satisfaction coefficient ω is as follows: ω=α1*ω1'+α2*ω2' It should be noted that α1 and α2 are preset proportional coefficients of ω1′ and ω2′, and 0<α1≤α2<1, and α1+α2=1.

8. The relay protection device setting calibration system according to claim 5, characterized in that: The first adjustment unit of the fixed value calibration model is electrically connected to the user satisfaction calculation unit and the fixed value calibration model construction module, and is used to obtain the user satisfaction coefficient ω calculated by the user satisfaction calculation unit, and transmit the user satisfaction coefficient ω to the fixed value calibration model construction module to adjust the model parameters of the fixed value calibration model. The specific adjustment formula is as follows: l new =λ old -s*th-ω*t It should be noted that λ old represents the old parameters of the fixed-value calibration model, λ new represents the new parameters of the fixed-value calibration model, σ represents the learning rate, which is used to control the step size of parameter update, θ represents the gradient of the objective function to the model parameters, and τ represents the gradient of the user feedback to the model parameters.

9. The relay protection device setting calibration system according to claim 5, characterized in that: The second adjustment unit of the fixed value calibration model is electrically connected to the user satisfaction calculation unit and the fixed value calibration model construction module, and is used to obtain the user satisfaction coefficient ω calculated by the user satisfaction calculation unit, and compare the user satisfaction coefficient ω with a preset threshold value TTH. When the user satisfaction coefficient ω is greater than the preset threshold value TTH, the user satisfaction coefficient ω in the training set and the test set is eliminated. n-m ', x n-m+1 ', ..., x n-1 'data set, use other data sets that satisfy the user to calculate the objective function of the fixed value calibration model; It should be noted that m is a natural number greater than 1.

10. A method for calibrating a relay protection device setting value, used in the relay protection device setting value calibration system according to any one of claims 1 to 9, characterized in that: The steps include: S1, real-time collection of equipment operation data and relay protection device setting data; S2, processing the operating data and the fixed value data, and dividing the processed results into a training set, a test set and a calibration set; S3, training a fixed-value calibration model according to the training set, and performing an evaluation test on the trained fixed-value calibration model according to the test set; S4, processing the calibration centralized data according to the fixed value calibration model to obtain adjustment data, and transmitting the adjustment data to the relay protection device for fixed value calibration; S5, collect the user's calibration accuracy score ω1 and the equipment operation effect score ω2, generate the user satisfaction coefficient ω, and adjust the parameters and objective function of the fixed value calibration model according to the user satisfaction coefficient ω.

Images