Capacitive voltage transformer metering error evaluation method and system, evaluation terminal and storage medium

Abstract

The invention provides a capacitor voltage transformer metering error evaluation method and system, an evaluation terminal and a storage medium, and the method comprises the steps: obtaining historical data of a voltage transformer, and analyzing time sequence data; calculating to obtain a first characteristic value according to the relationship between the time sequence data and the period; calculating a second characteristic value based on the influence of the stripping temperature on the error; calculating a relation parameter between the residual component and historical data to obtain a third characteristic value; importing the first characteristic value, the second characteristic value and the third characteristic value into a deep belief network training model for training to obtain a mutual inductor error prediction model; and converting the operation data of the voltage transformer in the to-be-detected time period into a first characteristic value, a second characteristic value and a third characteristic value, importing the first characteristic value, the second characteristic value and the third characteristic value into the transformer error prediction model for prediction, and outputting a voltage transformer error prediction value. According to the method, the deep belief network is improved, real-time real data is introduced through the gating function, the problem that the gradient of the original network disappears is solved, and the accuracy is improved.

Description

Background Art

[0002] As an important measuring device in power system, capacitor voltage transformer is widely used in voltage measurement and power metering. Its accuracy has a great impact on power grid operation, power metering and power quality analysis.

[0003] It is crucial to accurately evaluate the measurement error of capacitor voltage transformers. Traditionally, this evaluation usually relies on manual offline experiments and theoretical derivation, which is not only time-consuming and labor-intensive, but also limited by the control of the experimental environment and the assumptions of the theoretical model, and may not truly reflect the performance of the equipment in actual operation.

[0004] With the advancement of data acquisition technology and the development of data analysis algorithms, data-driven evaluation methods have gradually become mainstream in power system monitoring and analysis. For the error evaluation of capacitor voltage transformers, the online detection method has obvious advantages: the online detection method can evaluate based on the real-time collected capacitor voltage transformer output data, and reflect the actual performance and error of the equipment in real time. This real-time nature makes the evaluation results more representative and reliable.

[0005] Traditional manual offline evaluation is often limited by the complexity of controlling experimental conditions and environment, while the online detection method can be evaluated in a complex operating environment and adapt to the performance changes of capacitive voltage transformers under different working conditions.

[0006] At present, most of the capacitive voltage transformers are used in online detection, which can obtain voltage data in real time. However, since the use of capacitive voltage transformers is affected by the external environment and the internal operation of the system, such as voltage waveform characteristics, frequency response, temperature effects, etc., it is easy to cause errors to exceed the range and affect the accuracy of the detected voltage data. At present, it is impossible to predict the error state of the detection value within a certain period of time, nor is it possible to comprehensively evaluate the working state of the transformer. Summary of the invention

[0007] The present invention provides a method for evaluating the measurement error of a capacitive voltage transformer. The method uses three related characteristic quantities in combination with an improved deep belief network to perform predictive analysis on real-time real data, thereby being able to determine the error state of the current detection value, thereby improving the detection accuracy.

[0008] Methods include:

[0009] S101: Obtain historical data of the voltage transformer and parse out time series data;

[0010] S102: Calculate and obtain a first eigenvalue according to the relationship between the time series data and the period;

[0011] S103: stripping the influence of temperature on the error, calculating the second eigenvalue;

[0012] S104: Calculate the relationship parameter between the residual component and the historical data to obtain the third eigenvalue;

[0013] S105: importing the first eigenvalue, the second eigenvalue, and the third eigenvalue into a deep belief network training model for training to obtain a mutual inductor error prediction model;

[0014] S106: Convert the historical data of the voltage transformer at the time point to be tested in the previously preset time period into the first eigenvalue, the second eigenvalue, and the third eigenvalue, import the temperature, humidity, and secondary voltage of the voltage transformer at the time point to be tested into the transformer error prediction model for prediction, and output the error prediction value of the voltage transformer.

[0015] It should be further explained that, in step S101, the temperature data M, humidity data H and secondary voltage data U of the voltage transformer within a preset historical time period are obtained;

[0016] Decompose the time series data Q into trend component, period component and residual component, that is

[0017] Q v =S v +T v +R v ;

[0018] v is the data point sequence number (1≤v≤n), T v , S v , R v They are the trend component, periodic component and residual component at time v respectively.

[0019] It should be further explained that step S101 also includes: defining T v (k) , S v (k) is the trend component and period component at the end of the k-1th time in the inner cycle. At the initial time, T v (k) =0;

[0020] Perform the detrending process and subtract the trend component of the previous round result, Y v '=Y v -T v (k) ;

[0021] Using LOESS (q = n n(s) ,d=1) Perform regression operation on each subsequence and extend it forward and backward by one cycle;

[0022] The smoothed results form a temporary periodic time series, denoted as C v (k+1) ,v=-n (p) +1,…,-N+n (p) ;

[0023] Perform low-throughput filtering on the periodic subsequence, and the resulting sequence is L v (k+1) ,v=1,…,N;

[0024] Remove the trend of the smoothed periodic subseries: S v (k+1) =C v (k+1) -L v (k+1) ; Subtract the periodic component and perform trend smoothing to obtain the trend component T v (k+1) .

[0025] It should be further explained that in step S102, the temperature data M, humidity data H and secondary voltage data U are subjected to fast Fourier transform, and the inverse of the main frequency is recorded as PM, PH, and PU respectively, and the time series data S is subjected to fast Fourier transform, and the inverse of the main frequency is recorded as PS, and the first eigenvalue F is calculated by the following formula: 1 ;

[0026]

[0027] It should be further explained that in S103, the temperature affects the error ΔS (M) Calculation method:

[0028]

[0029] Where B is the rated load, is the initial constant, a c is the temperature coefficient, Δm is the difference between the temperature at the measuring point and 20°C, ω n is the rated angular frequency; I 1 is the primary current;

[0030] Time series of trend component T affected by stripping temperature b The calculation method is:

[0031] T b =T-ΔT (M) ;

[0032] The second eigenvalue F 2 The calculation method is:

[0033]

[0034] Δm' is the difference between the real-time measurement point temperature and 20°C, S y The current forecast value.

[0035] It should be further explained that the third eigenvalue F 3 The calculation method is:

[0036]

[0037] A 1 is the relationship parameter between the residual component R and the temperature data M, A 2 is the relationship parameter between the residual component R and the humidity data H, A 3 It is the relationship parameter between the residual component R and the secondary voltage data U.

[0038] It should be further explained that step S105 also includes: l 、Humidity data to be measured h l And the secondary voltage data u to be measured l Import the deep belief network training model for training, and use the error value of the preset historical time period as the output of the output layer;

[0039] The first eigenvalue F 1 , the second eigenvalue F 2 , the third eigenvalue F 3 As the explicit layer of the deep belief network training model, the four elements of the hidden layer are denoted as D t-1 , the temperature data to be measured m l 、Humidity data to be measured h l And the secondary voltage data u to be measured l Perform the following operations:

[0040] i t =σ(W i ·[m l ,h l ,u l ]+b i );

[0041]

[0042] σ represents the sigmoid activation function, b i 、b D Represents i t and The threshold vector of , tanh represents the Tanh function;

[0043] Parameter i t and parameters The calculation relationship is:

[0044]

[0045] D t To form the fusion layer and output as the combination.

[0046] The present application also provides a capacitive voltage transformer measurement error assessment system, the system comprising: a data acquisition and analysis module, a characteristic value calculation module, a model training module and an error prediction module;

[0047] A data acquisition and analysis module is used to acquire historical data of voltage transformers and parse out time series data;

[0048] The eigenvalue calculation module is used to calculate the first eigenvalue according to the relationship between the time series data and the period;

[0049] Based on the effect of the peeling temperature on the error, the second eigenvalue is calculated;

[0050] Calculate the relationship parameters between the residual component and the historical data to obtain the third eigenvalue;

[0051] A model training module, used for importing the first eigenvalue, the second eigenvalue and the third eigenvalue into a deep belief network training model for training to obtain a mutual inductor error prediction model;

[0052] The error prediction module is used to convert the operating data of the voltage transformer during the test time period into the first eigenvalue, the second eigenvalue, and the third eigenvalue, and import the transformer error prediction model for prediction, and output the voltage transformer error prediction value.

[0053] According to another embodiment of the present application, an evaluation terminal is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the capacitor voltage transformer metering error evaluation method when executing the program.

[0054] According to another embodiment of the present application, a storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method for evaluating the measurement error of a capacitor voltage transformer are implemented.

[0055] It can be seen from the above technical solutions that the present invention has the following advantages:

[0056] The capacitive voltage transformer measurement error assessment method provided in this application can more accurately capture the complex variation law of voltage transformer error by comprehensively considering time series data, temperature influence, residual components, etc., and using deep belief network (DBN) for training, thereby improving the accuracy of error prediction. The deep belief network used in this application can learn high-level abstract features of data through multi-layer nonlinear transformation, which enables the model to maintain good prediction performance when facing new and unseen data, and enhances the generalization ability of the model.

[0057] The capacitance voltage transformer measurement error assessment method realizes automation from data acquisition, feature extraction to error prediction, greatly reducing manual intervention and improving assessment efficiency. At the same time, by stripping away the direct impact of temperature on the error, the contribution of time changes, system status, etc. to the error can be more clearly seen, and the source of the error can be accurately identified, providing a basis for subsequent error correction and optimization.

[0058] The method of the present application can receive the operating data of the voltage transformer in real time, and make error prediction based on the data, thereby realizing dynamic monitoring of the transformer error. This is of great significance for ensuring the stable operation of the power system and timely discovering and handling potential problems.

[0059] During the evaluation process, this application not only considers the time series data of the voltage transformer, but also integrates external information such as temperature and humidity, realizing the fusion analysis of multi-source data, which helps to more comprehensively evaluate the working status of the transformer. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] In order to more clearly illustrate the technical solution of the present invention, the accompanying drawings required for use in the description will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0061] Figure 1 The present invention is a flow chart of an embodiment of a method for evaluating a measurement error of a capacitor voltage transformer;

[0062] Figure 2 It is a flow chart of the method for evaluating the measurement error of capacitor voltage transformer;

[0063] Figure 3 To improve the DBN network diagram;

[0064] Figure 4 Schematic diagram of the capacitor voltage transformer measurement error evaluation system. DETAILED DESCRIPTION

[0065] The following will describe in detail the capacitance voltage transformer measurement error evaluation method involved in the present application in conjunction with the accompanying drawings.

[0066] In order to facilitate understanding of the method provided in the embodiment of the present application, the following points are first explained: First, in the embodiment of the present application, "used for indication" may include direct indication and indirect indication, and may also include explicit indication and implicit indication. The information indicated by a certain information is called information to be indicated. In the specific implementation process, there are many ways to indicate the information to be indicated, such as but not limited to, the information to be indicated can be directly indicated, such as the information to be indicated itself or the index of the information to be indicated. The information to be indicated can also be indirectly indicated by indicating other information, wherein there is an association relationship between the other information and the information to be indicated. It is also possible to indicate only a part of the information to be indicated, while the other parts of the information to be indicated are known or agreed in advance. For example, it is also possible to implement the indication of the information to be indicated by means of a pre-agreement (such as a protocol provision) on whether a certain information element exists, thereby reducing the indication overhead to a certain extent.

[0067] In the embodiments of the present application, "at least one" refers to one or more, and "plurality" refers to two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone, where A and B can be singular or plural. The character " / " generally indicates that the previous and next associated objects are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b and c can represent: a, or b, or c, or a and b, or a and c, or b and c, or a, b and c, where a, b, c can be single or multiple.

[0068] In the embodiments of the present application, descriptions such as "when...", "in the case of...", "if" and "if" all refer to making corresponding processing on the voltage transformer metering error evaluation under certain objective circumstances. They are not limiting conditions, and do not require that the error evaluation must have a judgment action when it is implemented, nor do they mean the existence of other limitations.

[0069] like Figure 1The flowchart of the preferred embodiment of the capacitor voltage transformer measurement error evaluation method of the present invention is shown. The capacitor voltage transformer measurement error evaluation method is applied to one or more evaluation terminals, and the evaluation terminal is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes but is not limited to microprocessors, application specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), digital signal processors (DSP), embedded devices, etc.

[0070] The evaluation terminal is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workbenches, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The evaluation terminal can also represent various forms of mobile devices, such as personal digital processing, cellular phones, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the embodiments of the present application described and / or required herein.

[0071] 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.

[0072] See also Figure 1 and Figure 2 The figure is a flow chart of a method for evaluating a capacitance voltage transformer measurement error in a specific embodiment, the method comprising:

[0073] S101: Acquire historical data of a voltage transformer and parse out time series data.

[0074] In some embodiments, the historical data of voltage transformer measurement can be stored in a database. In addition to storing the data measured by the voltage transformer, the environmental information of the voltage transformer, such as temperature information and humidity information, can also be included, because the temperature and humidity of the environment will also affect the measurement error. Of course, other environmental information can also be considered to affect the measurement error according to actual needs. For each measurement of the voltage transformer, the data stored includes the corresponding voltage data, and the environmental information at the measurement time point can also be involved.

[0075] In an exemplary embodiment, information about the voltage transformer can be obtained for a period of time such as the past month or the past two weeks, and the specific time period is not limited here. The time series data of the capacitive voltage transformer detected by the offline detector is denoted as Q, and the number of historical data obtained is n. Historical data may include temperature data, humidity data, and secondary voltage data, denoted as M, H, and U, respectively. The real-time measurement point temperature m, real-time humidity h, and real-time secondary voltage indication u of the voltage transformer to be tested can also be obtained. The STL method is used to decompose the time series data Q of the preset time period into three components, namely, trend component, periodic component, and residual component, that is,

[0076] Qv=Sv+Tv+-Rv

[0077] v is the data point sequence number (1≤v≤n), T v , S v , R v They are the trend component, periodic component and residual component at time v respectively.

[0078] In the embodiment of the present application, it is assumed that T v (k) , S v (k) is the trend component and period component at the end of the k-1th time in the inner cycle. At the initial time, T v (k) = 0. The specific implementation process of the STL method is as follows:

[0079] S1001: Execute the detrending process, subtract the trend component of the result in the previous inner loop, Y v '=Y v -T v (k) .

[0080] S1002: Smoothing the periodic subsequence.

[0081] In some embodiments of the present application, LOESS (q = n n(s) , d=1) regress each subsequence (LOESS refers to local weighted regression, n(s) is the LOESS smoothing parameter, q is the data subset acquisition range, and d is the order in the local regression), and extend it forward and backward by one period respectively; the smoothing results constitute a temporary periodic time series, denoted as C v (k+1) ,v=-n (p) +1,…,-N+n (p) .

[0082] S1003: Filter low throughput of periodic subsequences.

[0083] It should be noted that, based on step S1002, the result sequence C is realizedv (k+1) Perform a sliding average of length n(p). n(p) is the number of samples in one cycle. Then perform LOESS (q = n n(l) ,d=1) regression processing.

[0084] n(s) is the LOESS smoothing parameter, and the resulting sequence L v (k+1) ,v=1,…,N。

[0085] S1004: Remove the trend of the smoothed periodic subsequence: S v (k+1) =C v (k+1) -L v (k+1) .

[0086] S1005: Based on the periodic component, Y v '=Y v –S v (k+1) Perform de-cycle processing.

[0087] S1006: Perform LOESS (q=n n(t) ,d=1) Regression trend smoothing process to obtain trend component T v (k+1) .

[0088] In this way, the trend component time series, period component time series, and residual component time series after the mutual inductor error time series is decomposed are obtained through the above steps, which are denoted as S, T, and R.

[0089] S102: Calculate and obtain a first eigenvalue based on the relationship between the time series data and the period.

[0090] According to the embodiment of the present application, the temperature data M, the humidity data H and the secondary voltage data U are subjected to fast Fourier transformation, and the inverse of the main frequency is recorded as PM, PH, and PU respectively. The time series data S is subjected to fast Fourier transformation, and the inverse of the main frequency is recorded as PS. According to the influence of the M, H, and U cycles on the S cycle, the first eigenvalue F is calculated by the following formula: 1 ;

[0091]

[0092] Simplify to get

[0093]

[0094] S103: Calculate a second eigenvalue based on the effect of the peeling temperature on the error.

[0095] In some embodiments, the effect of the peeling temperature T on the error ΔS (M) Calculation formula:

[0096]

[0097] Where B is the rated load, is an initial constant; ac is the temperature coefficient; Δm is the difference between the temperature of the measuring point and 20℃, ωn is the rated angular frequency; I1 is the primary current.

[0098] In this embodiment, the trend component time series T without the influence of temperature is b ,

[0099] T b =T-ΔT (M) .

[0100] Furthermore, for S b The data is linearly fitted using the least squares method, and the current predicted value Sy is added to the temperature influence value of the real-time temperature to calculate the second eigenvalue F2.

[0101]

[0102] Δm' is the difference between the real-time measurement point temperature and 20°C.

[0103] S104: Calculate the relationship parameters between the residual component and the historical data to obtain the third eigenvalue.

[0104] In some embodiments, relationship parameters between the residual component R and the temperature data M, the humidity data H, and the secondary voltage data U are calculated respectively.

[0105] A 1 It is the relationship parameter between the residual component R and the temperature data M.

[0106]

[0107] A 2 It is the relationship parameter between the residual component R and the humidity data H.

[0108]

[0109] A 3 It is the relationship parameter between the residual component R and the secondary voltage data U.

[0110]

[0111] Count() is a counting symbol. The calculation formula for the third feature F3 is as follows:

[0112]

[0113] S105: Importing the first eigenvalue, the second eigenvalue, and the third eigenvalue into a deep belief network training model for training to obtain a mutual inductor error prediction model.

[0114] In the embodiments of the present application, Figure 3 As shown, the temperature data to be measured within the preset historical time period m l 、Humidity data to be measured h l And the secondary voltage data u to be measured l Import the deep belief network training model for training, and use the error value of the preset historical time period as the output of the output layer.

[0115] The first eigenvalue F 1 , the second eigenvalue F 2 , the third eigenvalue F 3 As the explicit layer of the deep belief network training model, the four elements of the hidden layer are denoted as D t-1 , the temperature data to be measured m l 、Humidity data to be measured h l And the secondary voltage data u to be measured l Perform the following operations:

[0116]

[0117]

[0118] σ represents the sigmoid activation function, b i , b D Respectively represent i t and The threshold vector of , tanh represents the Tanh function;

[0119] Parameter i t and parameters The calculation relationship is:

[0120]

[0121] D t To form the fusion layer and output as the combination.

[0122] S106: Convert the historical data of the voltage transformer at the time point to be tested in the previously preset time period into the first eigenvalue, the second eigenvalue, and the third eigenvalue, import the temperature, humidity, and secondary voltage of the voltage transformer at the time point to be tested into the transformer error prediction model for prediction, and output the error prediction value of the voltage transformer.

[0123] Optionally, the historical data of the voltage transformer one month before the current day is converted into the first characteristic value, the second characteristic value, and the third characteristic value, and the temperature, humidity, and secondary voltage of the voltage transformer on the day to be tested are used as m l 、h l 、u l The transformer error prediction model is imported for prediction, and the output value is the voltage transformer error prediction value.

[0124] The capacitive voltage transformer measurement error assessment method provided in this application can more accurately capture the complex variation law of voltage transformer error by comprehensively considering time series data, temperature influence, residual components, etc., and using deep belief network (DBN) for training, thereby improving the accuracy of error prediction. The deep belief network used in this application can learn high-level abstract features of data through multi-layer nonlinear transformation, which enables the model to maintain good prediction performance when facing new and unseen data, and enhances the generalization ability of the model.

[0125] The capacitance voltage transformer measurement error assessment method realizes automation from data acquisition, feature extraction to error prediction, greatly reducing manual intervention and improving assessment efficiency. At the same time, by stripping away the direct impact of temperature on the error, the contribution of time changes, system status, etc. to the error can be more clearly seen, and the source of the error can be accurately identified, providing a basis for subsequent error correction and optimization.

[0126] The method of the present application can receive the operating data of the voltage transformer in real time, and make error prediction based on the data, thereby realizing dynamic monitoring of the transformer error. This is of great significance for ensuring the stable operation of the power system and timely discovering and handling potential problems.

[0127] During the evaluation process, this application not only considers the time series data of the voltage transformer, but also integrates external information such as temperature and humidity, realizing the fusion analysis of multi-source data, which helps to more comprehensively evaluate the working status of the transformer.

[0128] In some specific embodiments, since CVT is currently susceptible to damage and it is necessary to understand the CVT status in real time, but when there is no condition for offline detection of the transformer, the real-time error can be calculated based on the real-time environmental data and historical data, thus getting rid of the traditional method of offline detection of CVT and improving the accuracy of error assessment.

[0129] The results of evaluating the CVT error (ratio error) using the method of the present invention are shown in Table 1.

[0130] Table 1

[0131] Serial number CVT Error Evaluation Results Offline ground truth results 1 0.05254％ 0.06012％ 2 0.02841％ 0.02744％ 3 0.03457％ 0.03484％ 4 0.05431％ 0.05547％ 5 0.08101％ 0.07980％ 6 0.04233％ 0.04698％ 7 0.05457％ 0.04634％ 8 0.02857％ 0.02865％ 9 0.01741％ 0.02963％

[0132] It can be seen from Table 1 that the error evaluation deviation of this method is small. To further quantify the accuracy, the root mean square error is used to represent it.

[0133]

[0134] The root mean square error is small, which means that the transformer error evaluation value obtained by the algorithm of the present invention is very close to the true value and can be used in practical applications.

[0135] The following is an embodiment of a capacitor voltage transformer metering error evaluation system provided by an embodiment of the present disclosure. The system and the capacitor voltage transformer metering error evaluation method of the above-mentioned embodiments belong to the same inventive concept. For details not described in detail in the embodiment of the capacitor voltage transformer metering error evaluation system, reference can be made to the embodiment of the above-mentioned capacitor voltage transformer metering error evaluation method.

[0136] A mobile terminal implementing various embodiments of the present invention will now be described with reference to the accompanying drawings. In the subsequent description, suffixes such as "module", "component" or "unit" used to represent elements are used only to facilitate the description of the embodiments of the present invention and have no specific meaning in themselves. Therefore, "module" and "component" can be used interchangeably.

[0137] like Figure 4 As shown, the system includes: a data acquisition and analysis module, a eigenvalue calculation module, a model training module and an error prediction module.

[0138] The data acquisition and analysis module is used to obtain the historical data of the voltage transformer and parse out the time series data.

[0139] The eigenvalue calculation module is used to calculate the first eigenvalue according to the relationship between the time series data and the period; calculate the second eigenvalue based on the influence of the stripping temperature on the error; and calculate the relationship parameters between the residual component and the historical data to obtain the third eigenvalue.

[0140] The model training module is used to import the first eigenvalue, the second eigenvalue and the third eigenvalue into the deep belief network training model for training to obtain the mutual inductor error prediction model.

[0141] The error prediction module is used to convert the operating data of the voltage transformer during the test time period into the first eigenvalue, the second eigenvalue, and the third eigenvalue, and import the transformer error prediction model for prediction, and output the voltage transformer error prediction value.

[0142] The capacitive voltage transformer metering error evaluation system involved in this embodiment is a unit and algorithm step of each example described in combination with the embodiments disclosed herein, and can be implemented by electronic hardware, computer software or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to the function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0143] It will be appreciated by those skilled in the art that various aspects of the method for evaluating the measurement error of a capacitor voltage transformer can be implemented as a system, method or program product. Therefore, various aspects of the present disclosure can be specifically implemented in the following forms, namely: a complete hardware implementation, a complete software implementation (including firmware, microcode, etc.), or a combination of hardware and software, which can be collectively referred to as "circuit", "module" or "system" here.

[0144] In the storage medium provided in the present application, a program product capable of implementing the above method of the present specification is stored. In some possible implementations, various aspects of the present disclosure may also be implemented in the form of a program product, which includes a program code, and when the program product is run on a terminal device, the program code is used to enable the terminal device to execute the steps according to various exemplary implementations of the present disclosure described in the above "Exemplary Method" section of the present specification.

[0145] The storage medium can adopt any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium can be, for example, but not limited to, a system, device or device of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination of the above. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0146] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for evaluating measurement error of a capacitor voltage transformer, characterized in that: Methods include: S101: Obtain historical data of the voltage transformer and parse out time series data; S102: Calculate and obtain a first eigenvalue according to the relationship between the time series data and the period; S103: stripping the influence of temperature on the error, calculating the second eigenvalue; S104: Calculate the relationship parameter between the residual component and the historical data to obtain the third eigenvalue; S105: importing the first eigenvalue, the second eigenvalue, and the third eigenvalue into a deep belief network training model for training to obtain a mutual inductor error prediction model; S106: Convert the historical data of the voltage transformer at the time point to be tested in the previously preset time period into the first eigenvalue, the second eigenvalue, and the third eigenvalue, import the temperature, humidity, and secondary voltage of the voltage transformer at the time point to be tested into the transformer error prediction model for prediction, and output the error prediction value of the voltage transformer.

2. The method for evaluating the measurement error of a capacitor voltage transformer according to claim 1, characterized in that: In step S101, temperature data M, humidity data H and secondary voltage data U of the voltage transformer within a preset historical time period are obtained; Decompose the time series data Q into trend component, period component and residual component, that is Q v =S v +T v +R v ; v is the data point sequence number (1≤v≤n), T v , S v , R v They are the trend component, periodic component and residual component at time v respectively.

3. The method for evaluating the measurement error of a capacitor voltage transformer according to claim 2, characterized in that: Step S101 also includes: defining T v (k) , S v (k) is the trend component and period component at the end of the k-1th time in the inner cycle. At the initial time, T v (k) =0; Perform the detrending process and subtract the trend component of the previous round result, Y v '=Y v -T v (k) ; Using LOESS (q = n n(s) ,d=1) Perform regression operation on each subsequence and extend it forward and backward by one cycle; The smoothed results form a temporary periodic time series, denoted as C v (k+1) ,v=-n (p) +1,…,-N+n (p) ; Perform low-throughput filtering on the periodic subsequence, and the resulting sequence is L v (k+1) ,v=1,…,N; Remove the trend of the smoothed periodic subseries: S v (k+1) =C v (k+1) -L v (k+1) ; Subtract the periodic component and perform trend smoothing to obtain the trend component T v (k+1) .

4. The method for evaluating the measurement error of a capacitor voltage transformer according to claim 1, characterized in that: In step S102, the temperature data M, humidity data H and secondary voltage data U are subjected to fast Fourier transformation, and the inverse of the main frequency is recorded as PM, PH and PU respectively. The time series data S is subjected to fast Fourier transformation, and the inverse of the main frequency is recorded as PS. The first eigenvalue F1 is calculated by the following formula; 5. The method for evaluating capacitance voltage transformer measurement error according to claim 1, characterized in that: In S103, the temperature affects the error ΔS (M) Calculation method: Among them, B is the rated load, is the initial constant, a c is the temperature coefficient, Δm is the difference between the temperature at the measuring point and 20°C, ω n is the rated angular frequency; I1 is the primary current; The trend component time series T of the stripping temperature effect b The calculation method is: T b =T-ΔT (M) ; The second eigenvalue F2 is calculated as: Δm' is the difference between the real-time measurement point temperature and 20°C, S y The current forecast value.

6. The method for evaluating capacitance voltage transformer measurement error according to claim 1, characterized in that: The third eigenvalue F3 is calculated as: A1 is the relationship parameter between the residual component R and the temperature data M, A2 is the relationship parameter between the residual component R and the humidity data H, and A3 is the relationship parameter between the residual component R and the secondary voltage data U.

7. The method for evaluating capacitance voltage transformer measurement error according to claim 1, characterized in that: Step S105 also includes: storing the temperature data m to be measured within a preset historical time period l 、Humidity data to be measured h l And the secondary voltage data u to be measured l Import the deep belief network training model for training, and use the error value of the preset historical time period as the output of the output layer; The first eigenvalue F1, the second eigenvalue F2, and the third eigenvalue F3 are used as the explicit layer of the deep belief network training model, and the four elements of the hidden layer are recorded as D t-1 , the temperature data to be measured m l 、Humidity data to be measured h l And the secondary voltage data u to be measured l Perform the following operations: i t =σ(W i ·[m l ,h l ,u l ]+b i ); σ represents the sigmoid activation function, b i 、b D Represents i t and The threshold vector of , tanh represents the Tanh function; Parameter i t and parameter D t The calculation relationship is: D t To form the fusion layer and output as the combination.

8. A capacitive voltage transformer measurement error evaluation system, characterized in that: The system is used to implement the capacitance voltage transformer measurement error evaluation method according to any one of claims 1 to 7; the system comprises: a data acquisition and analysis module, a characteristic value calculation module, a model training module and an error prediction module; A data acquisition and analysis module is used to acquire historical data of voltage transformers and parse out time series data; The eigenvalue calculation module is used to calculate the first eigenvalue according to the relationship between the time series data and the period; Based on the effect of the peeling temperature on the error, the second eigenvalue is calculated; Calculate the relationship parameters between the residual component and the historical data to obtain the third eigenvalue; A model training module, used for importing the first eigenvalue, the second eigenvalue and the third eigenvalue into a deep belief network training model for training to obtain a mutual inductor error prediction model; The error prediction module is used to convert the operating data of the voltage transformer during the test time period into the first eigenvalue, the second eigenvalue, and the third eigenvalue, and import the transformer error prediction model for prediction, and output the voltage transformer error prediction value.

9. An evaluation terminal, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps of the method for evaluating the measurement error of a capacitor voltage transformer as claimed in any one of claims 1 to 7 are implemented.

10. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for evaluating a capacitance voltage transformer measurement error as claimed in any one of claims 1 to 7 are implemented.

Images