Method and system for eliminating the influence of vibration on optical current transformer based on waveform recognition

Abstract

The present application relates to the field of photoelectric technology, and particularly relates to a method and system for eliminating the vibration influence of an optical current transformer based on waveform identification. The method comprises the following steps: collecting demodulation waveform characteristic quantities and vibration characteristic quantities of the optical current transformer as a sample set, and using a support vector machine supervised learning model to train the sample set to obtain a decision model. On the basis of an existing sensor, the decision model is used to identify the vibration characteristic quantities in the real-time collected demodulation waveform characteristic quantities of the optical current transformer, so that the current values affected by vibration can be effectively eliminated, and the current values of a specific sampling rate are obtained through interpolation resampling technology to meet the interval sampling of the current values of the specific sampling rate required by the back-end application. The method and system do not need to additionally increase the sensor, are suitable for open-loop square wave modulation algorithms and closed-loop modulation algorithms, the decision model is only related to the support vector, the calculation amount and memory requirement are effectively reduced, the deployment threshold is reduced, and the method and system have the value of popularization and application in field implementation and cost compression.

Description

Technical Field

[0001] This invention relates to the field of optoelectronic technology, and specifically to a method and system for eliminating the influence of vibration in optical current transformers based on waveform recognition. Background Technology

[0002] As a primary sensing device in a power system, current transformers play a crucial role in reflecting the status of primary equipment. Relay protection equipment and dispatch control both require current data. Therefore, current transformers need to operate stably and reliably. Optical current transformers (OCTs) offer advantages such as small size, light weight, simple insulation structure, and open secondary circuitry. Currently, most optical current transformers adopt the Sagnac structure, which offers good reciprocity. Their optical paths are completely symmetrical, with two polarized beams always propagating in two orthogonal modes on the same optical fiber, and the optical paths of the two beams are identical. They exhibit good frequency and transient characteristics, are passive at the primary end, have strong anti-interference capabilities, and are safe and environmentally friendly. They are also easy to digitize and are increasingly used in power systems.

[0003] Due to its good reciprocity, optical current transformers can effectively suppress the impact of low-frequency vibrations on the system, but their vibration suppression effect is limited on the side far from the sensing ring.

[0004] To address the aforementioned issues, the existing technical solutions include Chinese invention patent application document CN108287262, "Temperature and Vibration Feedback Compensation System and Measurement Method for All-Fiber Optic Current Transformer," which uses the addition of vibration and temperature sensors for temperature and vibration measurement, and then a computer corrects the measured current based on the vibration and temperature values; and Chinese invention patent application document CN109490602, "Anti-interference Method and Optical Current Transformer System for Optical Current Transformer," which uses the addition of a current detection module unaffected by vibration, and uses the measured value of the current detection module to replace the output current value of the optical current transformer when vibration occurs.

[0005] Existing research on eliminating the effects of vibration on optical current transformers requires the addition of additional sensors, which greatly increases the complexity of the optical current transformer measurement system and the difficulty of field application.

[0006] Therefore, we have invented a new method and system for eliminating the effects of vibration in optical current transformers. This method solves the technical problem that existing solutions require additional sensors to eliminate the effects of vibration, while ensuring the effective removal of current values ​​affected by vibration. This simplifies the workflow, saves costs, and lowers the deployment threshold.

[0007] Furthermore, this invention is applicable to both open-loop square wave modulation algorithms and closed-loop modulation algorithms. Interpolation resampling technology is used to obtain current values ​​sampled at equal intervals with a specific sampling rate that meet the requirements of backend applications.

[0008] It should be noted that the information disclosed in the background section above is only used to enhance the understanding of the background of the present invention. Summary of the Invention

[0009] The purpose of this invention is to provide a method and system for eliminating the vibration effects of optical current transformers based on waveform recognition, so as to solve the technical problem mentioned in the background art that existing solutions require additional sensors to eliminate vibration effects.

[0010] To achieve the above objectives, the present invention adopts the following technical solution:

[0011] According to one aspect of the present invention, a method for eliminating the influence of vibration in an optical current transformer based on waveform recognition is provided, comprising the following steps:

[0012] The demodulated waveform characteristics and vibration characteristics of the optical current transformer are collected as a sample set, where the demodulated waveform characteristics are the independent variables of the sample set and the vibration characteristics are the dependent variables of the sample set.

[0013] The sample set is trained with a supervised learning model to obtain a decision model for the demodulated waveform features and vibration features, wherein the kernel function is a linear kernel and the decision model function is a support vector.

[0014] Based on the decision model, the real-time demodulated waveform characteristics of the optical current transformer are input to obtain the real-time vibration characteristics.

[0015] The decision model is used to compare the real-time vibration characteristic values, and demodulation current values ​​with real-time vibration characteristic values ​​greater than 1 are removed to obtain the current values ​​that eliminate the vibration influence.

[0016] Based on the current value that eliminates the influence of vibration, interpolation and resampling are performed to obtain the output current value sampled at equal intervals with a specific sampling rate.

[0017] Optionally, the demodulated waveform of the optical current transformer is the electrical signal waveform of the transformer's photodetector.

[0018] Optionally, the demodulated waveform characteristics of the optical current transformer include, but are not limited to, one or more of the maximum value, minimum value, and harmonic content.

[0019] Optionally, the demodulated waveform characteristics or vibration characteristics of the optical current transformer include, but are not limited to, the current value.

[0020] Optionally, a method for supervising the training of a decision model for the demodulated waveform features and vibration features of the optical current transformer obtained by training the supervised learning model on the sample set includes, but is not limited to, support vector machines.

[0021] Optionally, the maximum value x1, minimum value x2, and harmonic content x3 of the demodulated waveform characteristics are related to the vibration characteristic x. i The relation is f(x) i = 0.3x1 + 0.2x2 + 0.9x3.

[0022] Optionally, the vibration characteristics include, but are not limited to, waveform characteristics that differ from those in the electrical signal waveforms output by conventional sensors or optical current transformers that are unaffected by vibration, as simulated through field operation and testing.

[0023] Optionally, the interpolation resampling includes, but is not limited to, Lagrange interpolation.

[0024] Optionally, the sampling rate of the interpolation resampling is the reciprocal of the time interval between two adjacent resampling interpolation times.

[0025] According to one aspect of the present invention, a system for eliminating the influence of vibration in an optical current transformer based on waveform recognition is provided, comprising the following modules:

[0026] The data acquisition module is used to acquire the demodulated waveform characteristics and vibration characteristics output by the optical current transformer.

[0027] The machine learning module is used to receive the demodulated waveform features and vibration features as a sample set, wherein the demodulated waveform features are the independent variables of the sample set and the vibration features are the dependent variables of the sample set; and to train a supervised learning model on the sample set to obtain a decision model for the demodulated waveform features and vibration features, wherein the kernel function type is a linear kernel and the decision model function is a support vector.

[0028] The data processing module is used to input real-time optical current transformer demodulation waveform characteristics based on the decision model to obtain real-time vibration characteristics.

[0029] The vibration processing module is used to compare the real-time vibration characteristic quantities through the decision model, remove demodulation current values ​​with real-time vibration characteristic quantities greater than 1, and obtain current values ​​that eliminate the vibration influence.

[0030] The signal processing module is used to perform interpolation resampling based on the current value that eliminates the influence of vibration, so as to obtain the output current value sampled at equal intervals with a specific sampling rate;

[0031] The real-time processing module has three sub-modules: data processing module, vibration processing module, and signal processing module, which are used to process real-time data.

[0032] As can be seen from the above technical solution, compared with the prior art, the present invention has at least the following advantages and positive effects:

[0033] 1. Based on existing sensors, vibration characteristics can be identified in the demodulated waveform of optical current transformers acquired in real time through modeling, which can effectively eliminate current values ​​affected by vibration.

[0034] 2. Obtain current values ​​sampled at equal intervals with a specific sampling rate that meet the requirements of backend applications through interpolation resampling technology;

[0035] 3. No additional sensors are required, making it more feasible and cost-effective for on-site implementation, and thus possessing certain promotional and application value;

[0036] 4. Applicable to both open-loop square wave modulation and closed-loop modulation algorithms;

[0037] 5. After training with support vector machines, there is no need to save the training samples. The final model is only related to the support vectors, which effectively reduces the amount of computation and memory requirements, and lowers the deployment threshold of this method and system. Attached Figure Description

[0038] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention. It is obvious that the drawings described below are merely some embodiments of this disclosure, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort.

[0039] Figure 1 This is a schematic flowchart of a method for eliminating the vibration influence of an optical current transformer based on waveform recognition, provided in the first embodiment of the present invention.

[0040] Figure 2 This is a schematic diagram of an optical current transformer structure provided in the second embodiment of the present invention;

[0041] Figure 3 This is a schematic diagram of an interpolation resampling method provided in the third embodiment of the present invention;

[0042] Figure 4 This is a system for eliminating the vibration effects of optical current transformers based on waveform recognition, provided in the fourth embodiment of the present invention.

[0043] in, Figure 2 The markings are explained below:

[0044] 2-Acquisition unit; 3-Fiber optic sensing ring; 4-Transmission fiber; 5-Primary conductor; 21-Light source; 22-Coupled; 23-Polarizer; 24-Phase modulator; 25-Delay fiber; 26-Photodetector; 27-Analog-to-digital converter; 28-Digital-to-analog converter; 29-Signal processor; 31-λ / 4 waveplate; 32-Sensing fiber; 33-Mirror;

[0045] Figure 4 The markings are explained below:

[0046] 401 - Data Acquisition Module; 402 - Machine Learning Module; 403 - Data Processing Module; 404 - Vibration Processing Module; 405 - Signal Processing Module; 406 - Real-time Processing Module. Detailed Implementation

[0047] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided to make the invention more comprehensive and complete, and to fully convey the concept of the exemplary embodiments to those skilled in the art.

[0048] Furthermore, the described features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. Numerous specific details are provided in the following description to give a full understanding of embodiments of the invention. However, those skilled in the art will recognize that the technical solutions of the invention can be practiced without one or more of the specific details, or other methods, components, apparatuses, steps, etc., can be employed. In other instances, well-known methods, apparatuses, implementations, or operations are not shown or described in detail to avoid obscuring various aspects of the invention.

[0049] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor devices and / or microcontroller devices.

[0050] The flowcharts shown in the accompanying drawings are merely illustrative and do not necessarily include all content and operations / steps, nor do they necessarily have to be performed in the described order. For example, some operations / steps can be broken down, while others can be combined or partially combined; therefore, the actual execution order may change depending on the specific circumstances.

[0051] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0052] Example 1

[0053] Figure 1 This invention illustrates a method for eliminating the vibration effects of optical current transformers based on waveform recognition, the specific steps of which include:

[0054] The demodulated waveform characteristics and vibration characteristics of the optical current transformer are collected as a sample set, where the demodulated waveform characteristics are the independent variables of the sample set and the vibration characteristics are the dependent variables of the sample set.

[0055] It should be noted that the demodulated waveform is the electrical signal waveform output by the optical current transformer, and the characteristic quantities include, but are not limited to, the maximum value, minimum value, and harmonic content as independent variables of the sample set. This method and system do not require additional sensors and are applicable to both open-loop square wave modulation algorithms and closed-loop modulation algorithms.

[0056] In this embodiment, by simulating on-site operation tests, the current waveforms output by a traditional sensor unaffected by vibration and an optical current transformer are compared. If they do not match, it can be determined whether vibration has affected the device. The demodulated waveform characteristics that may be caused by vibration are recorded, and the vibration characteristics and corresponding current values ​​are used as a sample set. After all the required data has been tested, the sample sets of demodulated waveform characteristics and vibration characteristics are created.

[0057] Furthermore, a supervised learning model is trained on the sample set to obtain a decision model for the demodulated waveform features and vibration features, wherein the kernel function type is a linear kernel and the decision model function is a support vector.

[0058] It should be noted that the demodulated waveform characteristics and vibration characteristics mentioned are all per-unit data. The decision model refers to the decision model for the vibration characteristics of the optical current transformer.

[0059] Furthermore, the decision model can be obtained by training the sample set using a support vector machine.

[0060] Support Vector Machines (SVMs) are a type of generalized linear classifier that performs binary classification of data using supervised learning. Their decision boundary is the hyperplane with the largest margin, calculated from the training samples. Their basic model is a linear classifier with the largest margin defined in the feature space; this maximum margin distinguishes it from the perceptron.

[0061] In this embodiment, the decision model is trained using a linearly separable support vector machine with a linear kernel function. The slack variables for the soft margin are the maximum value x1 and the minimum value x2, and the penalty parameter for both slack variables is 0.1.

[0062] After training with SVM, there is no need to save the training samples. The final model is only related to the support vectors, which effectively reduces the amount of computation and memory requirements, and lowers the deployment threshold of this method and system.

[0063] The optical current transformer collects real-time demodulated waveform characteristics that need to be judged: maximum value x1, minimum value x2, and harmonic content x3 as input values. Substitute these values ​​into the relationship between demodulated waveform characteristics and vibration characteristics obtained from the above decision model to calculate the corresponding vibration characteristics. Then, determine whether the demodulated waveform of the optical current transformer has been affected by vibration, and thus determine whether the demodulated current point needs to be removed.

[0064] Furthermore, when the vibration feature quantity is located in the positive plane, the vibration feature quantity is greater than 0, indicating that the demodulated waveform is affected by vibration; when the vibration feature quantity is located in the negative plane, it indicates that the demodulated waveform is not affected by vibration. The support vector offset of the vibration feature quantity obtained during training is 1. Therefore, when the vibration feature quantity is greater than 1, the current demodulated current value data is discarded; when the vibration feature quantity is less than or equal to 1, the current demodulated current value data is retained. Thus, the data obtained after discarding is the current data with the vibration effect eliminated.

[0065] In this embodiment, the maximum value x1, minimum value x2, harmonic content x3, and vibration characteristic quantity x are obtained from the demodulated waveform feature quantities through training. i The relation is f(x) i = 0.3x1 + 0.2x2 + 0.9x3.

[0066] The demodulated waveform characteristics acquired in real time by the optical current transformer are used as input values ​​and substituted into the relationship between the demodulated waveform characteristics and vibration characteristics obtained from the above decision model to calculate the corresponding vibration characteristics. When the vibration characteristic is greater than 1, the current demodulated current value data is discarded; when the vibration characteristic is less than or equal to 1, the current demodulated current value data is retained. Therefore, the data obtained after discarding is the current data after eliminating the influence of vibration.

[0067] The obtained current data, after eliminating the effects of vibration, is then interpolated and resampled.

[0068] Furthermore, the specific sampling rate required can be determined based on the needs of the backend application.

[0069] Furthermore, Lagrange interpolation can be used to resample non-uniform sampling points. Lagrange interpolation is a polynomial interpolation method that finds a polynomial that takes the observed values ​​at exactly each observation point; such a polynomial is called a Lagrange polynomial. Mathematically, Lagrange interpolation can provide a polynomial function that passes through several known points on a two-dimensional plane.

[0070] Interpolation resampling can be used to obtain current values ​​sampled at equal intervals with a specific sampling rate that meet the requirements of backend applications.

[0071] Example 2

[0072] Figure 2 An optical current transformer structure to which this invention can be applied is shown. This embodiment describes its specific working process and the application of this invention in this embodiment, including the following steps:

[0073] The optical current transformer consists of a data acquisition unit 2, an optical fiber sensing loop 3, and a transmission optical fiber 4. The function of the optical current transformer is to detect the current in the primary conductor 5 and output an electrical signal by measuring the cumulative effect of the magnetic field within the optical loop surrounding the current being measured, based on Ampere's law and the Faraday magneto-optical effect.

[0074] The driving circuit in signal processor 29 drives light source 21 to emit light. The light emitted by light source 21 reaches the first end of coupler 22, and coupler 22 guides the light to its third end; after passing through coupler 22, the light enters polarizer 23, generating linearly polarized light. The linearly polarized light is split into two incident orthogonal linearly polarized beams, which pass through phase modulator 24, delay fiber 25, and transmission fiber 4, reaching one end of λ / 4 waveplate 31 in fiber optic sensing ring 3. The other end of λ / 4 waveplate 31 of fiber optic sensing ring 3 outputs two orthogonal circularly polarized beams. The two orthogonal circularly polarized beams propagate along sensing fiber 32 of fiber optic sensing ring 3. Due to the Faraday magneto-optical effect, in sensing fiber 32, one of the two orthogonal circularly polarized beams propagates faster and the other propagates slower, thus generating a phase difference.

[0075] After the two orthogonal circularly polarized beams are reflected by the mirror 33 at the end of the sensing fiber 32, the polarization modes of the two orthogonal circularly polarized beams are interchanged due to the effect of the mirror. The left-hand circularly polarized beam becomes the right-hand circularly polarized beam, and the right-hand circularly polarized beam becomes the left-hand circularly polarized beam, and then returns along the original path.

[0076] When the two returning orthogonal circularly polarized beams return, the direction of the magnetic field of the primary current remains unchanged, while the propagation direction and polarization state of the two returning orthogonal circularly polarized beams change, so the phase difference produced by the Faraday effect doubles. After passing through the λ / 4 waveplate 31 of the fiber optic sensing ring 3 again, they become two returning orthogonal linearly polarized beams, and their polarization directions are interchanged during their relative propagation.

[0077] When the two returning orthogonally polarized beams return, they pass sequentially through the transmission fiber 4 and the phase modulator 24 to become combined polarized beams, which then reach the polarizer 23. The interference light signal returned by the polarizer 23 reaches the third end of the coupler 22 and then returns to the photodetector 26 through the second end of the coupler 22.

[0078] The photodetector 26 performs photoelectric conversion on the interference light signal and outputs an electrical signal. The analog-to-digital converter 27 converts the analog electrical signal output by the photodetector 26 into a digital electrical signal, and the signal processor 29 generates a fixed-frequency modulation signal or a superposition signal of a fixed-frequency modulation signal and a digital step wave. The digital-to-analog converter 28 receives the square wave signal or the superposition signal of the square wave signal and the digital step wave and converts it into an analog electrical signal, which is then applied to the phase modulator 24.

[0079] The signal processing circuit 29 receives the digital electrical signal output by the analog-to-digital converter 27 and demodulates the digital electrical signal to determine the measured current in the primary conductor 5 located in the optical fiber sensing ring 3 and outputs the demodulated waveform.

[0080] In this embodiment, the acquisition unit module acquires the demodulated waveform characteristics and vibration characteristics of the signal processing circuit 29. The data generated and acquired during the field operation test simulation will be recorded as a sample set, and a decision model for the demodulated waveform characteristics and vibration characteristics of the optical current transformer will be trained on the sample set.

[0081] Furthermore, when the primary conductor 5 is replaced with the conductor that actually needs to be measured, the data acquisition module acquires demodulated waveform feature quantities from the real-time demodulated waveform output by the signal processing circuit 29. The acquired real-time demodulated waveform feature quantities are then transmitted to the real-time processing module. Based on the decision model, the real-time demodulated waveform feature quantities are input to obtain real-time vibration feature quantities. The real-time vibration feature quantities are compared using the decision model, and the current demodulated current value is selected to be discarded or retained to obtain the current value that eliminates the vibration influence. Based on the current value that eliminates the vibration influence, interpolation resampling is performed to obtain the output current value with a specific sampling rate.

[0082] As described in this embodiment, by adopting the method and system for eliminating vibration effects of this invention, the demodulated current value affected by vibration can be effectively eliminated. Finally, the interpolation resampling technology ensures that the output sampled current waveform is equally spaced, which meets the requirements of the backend application. This method does not require additional sensors and is applicable to both open-loop square wave modulation algorithm and closed-loop modulation algorithm, and has certain promotion and application value.

[0083] Example 3

[0084] Figure 3An interpolation resampling method applicable to embodiments of the present invention is shown, the process specifically including the following steps:

[0085] Lagrange interpolation is a polynomial interpolation method that finds a polynomial that takes the observed values ​​at exactly the points observed; such a polynomial is called a Lagrange polynomial. Mathematically, Lagrange interpolation can provide a polynomial function that passes through several known points in a two-dimensional plane.

[0086] Specifically, the basis functions of the Lalange interpolation function are symmetric, and the interpolation result is independent of the order of the interpolation points, making it easy to implement in programming.

[0087] Furthermore, the resampling interpolation timing is related to the sampling rate, and the sampling rate and the sampling time interval are reciprocals of each other. The sampling rate depends on the requirements of the backend application.

[0088] In this embodiment, when the sampling rate is 50kHz, the interval between two adjacent resampling times is 20μs.

[0089] Figure 3 In the diagram, x0 is the reception time of the original sampled data y0, x1 is the reception time of the y1 value, x2 is the reception time of the y2 value, and x is the resampling time. The value corresponding to x can be calculated using the following formula:

[0090]

[0091] Figure 3 In the diagram, the sampled value corresponding to the resampling time y is x, x 丢弃0 To determine y in the decision model 丢弃0 The demodulated waveform value is an abnormal demodulation current value.

[0092] By using interpolation resampling technology, current values ​​sampled at equal intervals with a specific sampling rate can be obtained to meet the requirements of backend applications.

[0093] Example 4

[0094] Figure 4 This invention illustrates a system for eliminating the vibration effects of optical current transformers based on waveform recognition, which specifically includes the following modules:

[0095] The data acquisition module 401 is used to acquire the demodulated waveform characteristics and vibration characteristics output by the optical current transformer.

[0096] The machine learning module 402 is used to receive the collected demodulated waveform features and vibration features as a sample set, and to train the sample set to obtain a decision model for the demodulated waveform features and vibration features of the optical current transformer.

[0097] The data processing module 403 is used to input the real-time optical current transformer demodulation waveform characteristic quantity according to the decision model to obtain the real-time vibration characteristic quantity.

[0098] The vibration processing module 404 is used to judge the real-time vibration characteristic quantity, select to discard or retain the current demodulation current value, and obtain the current value that eliminates the vibration influence.

[0099] The signal processing module 405 is used to perform interpolation resampling based on the current value that eliminates the influence of vibration, so as to obtain the output current value sampled at equal intervals.

[0100] The real-time processing module 406 includes three sub-modules: a data processing module 403, a vibration processing module 404, and a signal processing module 405, which are used to process real-time data.

[0101] Furthermore, the demodulated waveform characteristics and vibration characteristics of the optical current transformer output acquired by the data acquisition module 401 include, but are not limited to, the demodulated waveform characteristics and vibration characteristics of the optical current transformer in field operation test simulation and real time.

[0102] Specifically, the workflow between the modules is as follows:

[0103] The data acquisition module 401 performs data acquisition, collecting the demodulated waveform characteristics and vibration characteristics output by the optical current transformer, and then transmits the collected demodulated waveform characteristics and vibration characteristics to the machine learning module 402.

[0104] The machine learning module 402 receives the demodulated waveform features and vibration features collected by the data acquisition module 401 and uses them as a sample set. It then trains a decision model for the demodulated waveform features and vibration features of the optical current transformer on the sample set.

[0105] Furthermore, in the machine learning module 402, the training method for the decision model of the demodulated waveform features and vibration features of the optical current transformer obtained by training the sample set adopts a supervised learning model, including but not limited to support vector machines.

[0106] Furthermore, when the conductor under test is the actual conductor that needs to be measured, the acquisition unit module 401 acquires the real-time demodulated waveform characteristics output by the optical current transformer, without needing to acquire the real-time vibration characteristics. The acquired real-time demodulated waveform characteristics will be transmitted to the real-time processing module 406 for further processing.

[0107] After receiving the real-time demodulated waveform characteristics from the optical current transformer output by the acquisition unit module 401, the real-time processing module 406 hands over the real-time demodulated waveform characteristics to the data processing module 403 for processing. The data processing module 403 inputs the real-time demodulated waveform characteristics into the decision model to obtain the real-time vibration characteristics. The vibration processing module 404 judges the real-time vibration characteristics and chooses to discard or retain the current demodulated current value.

[0108] Furthermore, when the real-time vibration characteristic value is greater than 1, the current demodulation current value is discarded; when the real-time vibration characteristic value is less than or equal to 1, the current demodulation current value is retained.

[0109] Furthermore, the real-time demodulated waveform characteristics of the optical current transformer are processed by the vibration processing module 404 to obtain a current value that eliminates the influence of vibration.

[0110] The signal processing module 405 receives the current value for eliminating vibration effects transmitted by the vibration processing module 404, and performs interpolation and resampling on the current value for eliminating vibration effects.

[0111] Furthermore, the basis functions of the Larlange interpolation function are symmetric, and the interpolation result is independent of the order of the interpolation points, making it easy to implement in programming.

[0112] Furthermore, the resampling interpolation timing is related to the sampling rate, and the sampling rate and the sampling time interval are reciprocals of each other. The sampling rate depends on the requirements of the backend application.

[0113] Furthermore, the current value that eliminates the influence of vibration is processed by the signal processing module 405 to obtain the output current value sampled at equal intervals.

[0114] Other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of the invention are indicated by the claims. It should be understood that the invention is not limited to the precise structures described above and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

Claims

1. A method for eliminating the influence of vibration in optical current transformers based on waveform recognition, characterized in that: The demodulated waveform characteristics and vibration characteristics of the optical current transformer are collected as a sample set, where the demodulated waveform characteristics are the independent variables of the sample set and the vibration characteristics are the dependent variables of the sample set. The sample set is trained with a supervised learning model to obtain a decision model for the demodulated waveform features and vibration features, wherein the kernel function is a linear kernel and the decision model function is a support vector. Based on the decision model, the real-time demodulated waveform characteristics of the optical current transformer are input to obtain the real-time vibration characteristics. The decision model is used to compare the real-time vibration characteristic values, and demodulation current values ​​with real-time vibration characteristic values ​​greater than 1 are removed to obtain the current values ​​that eliminate the vibration influence. Based on the current value that eliminates the influence of vibration, interpolation and resampling are performed to obtain the output current value sampled at equal intervals with a specific sampling rate.

2. The method for identifying and eliminating the vibration influence of the optical current transformer based on the waveform according to claim 1, characterized in that, The demodulated waveform of the optical current transformer is the electrical signal waveform of the transformer's photoelectric detector.

3. The method for eliminating the influence of optical current transformer vibration based on waveform recognition according to claim 1, characterized in that, The demodulated waveform characteristics of the optical current transformer include one or more of the following: maximum value, minimum value, and harmonic content.

4. The method for eliminating the influence of optical current transformer vibration based on waveform recognition according to claim 1, characterized in that, The demodulated waveform characteristics or vibration characteristics include the current value.

5. The method for eliminating the influence of optical current transformer vibration based on waveform recognition according to claim 1, characterized in that, The method for training the supervised learning model includes support vector machines.

6. The method for eliminating the influence of optical current transformer vibration based on waveform recognition according to claim 1, characterized in that, The relationship between the maximum value x1, minimum value x2, harmonic content x3 and the vibration characteristic xi in the demodulated waveform characteristic is f(xi) = 0.3x1 + 0.2x2 + 0.9x3.

7. The method for eliminating the influence of optical current transformer vibration based on waveform recognition according to claim 1, characterized in that, The vibration characteristics include waveform characteristics that differ from those of electrical signals output by conventional sensors or optical current transformers that are unaffected by vibration, as demonstrated through field operation and test simulations.

8. The method for eliminating the influence of optical current transformer vibration based on waveform recognition according to claim 1, characterized in that, The interpolation resampling method includes Lagrange interpolation.

9. The method for eliminating the influence of optical current transformer vibration based on waveform recognition according to claim 1, characterized in that, The sampling rate of the interpolation resampling is the reciprocal of the time interval between two adjacent resampling interpolation times.

10. A system for eliminating the influence of vibration in optical current transformers based on waveform recognition, characterized in that, The system includes: The data acquisition module is used to acquire the demodulated waveform characteristics and vibration characteristics output by the optical current transformer. The machine learning module is used to receive the demodulated waveform features and vibration features as a sample set, wherein the demodulated waveform features are the independent variables of the sample set and the vibration features are the dependent variables of the sample set; and to train a supervised learning model on the sample set to obtain a decision model for the demodulated waveform features and vibration features, wherein the kernel function type is a linear kernel and the decision model function is a support vector. The data processing module is used to input real-time optical current transformer demodulation waveform characteristics based on the decision model to obtain real-time vibration characteristics. The vibration processing module is used to compare the real-time vibration characteristic quantities through the decision model, remove demodulation current values ​​with real-time vibration characteristic quantities greater than 1, and obtain current values ​​that eliminate the vibration influence. The signal processing module is used to perform interpolation resampling based on the current value that eliminates the influence of vibration, so as to obtain the output current value sampled at equal intervals with a specific sampling rate; The real-time processing module has three sub-modules: data processing module, vibration processing module, and signal processing module, which are used to process real-time data.

Images