CN115542230B - Current transformer error estimation method and device based on diffusion model - Google Patents
Current transformer error estimation method and device based on diffusion model Download PDFInfo
- Publication number
- CN115542230B CN115542230B CN202211478921.XA CN202211478921A CN115542230B CN 115542230 B CN115542230 B CN 115542230B CN 202211478921 A CN202211478921 A CN 202211478921A CN 115542230 B CN115542230 B CN 115542230B
- Authority
- CN
- China
- Prior art keywords
- noise
- current transformer
- diffusion model
- power spectrum
- phase difference
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000009792 diffusion process Methods 0.000 title claims abstract description 70
- 238000000034 method Methods 0.000 title claims abstract description 45
- 238000009826 distribution Methods 0.000 claims abstract description 53
- 238000001228 spectrum Methods 0.000 claims abstract description 51
- 238000012549 training Methods 0.000 claims abstract description 16
- 239000013598 vector Substances 0.000 claims description 18
- 238000010586 diagram Methods 0.000 claims description 13
- 238000004590 computer program Methods 0.000 claims description 12
- 230000007246 mechanism Effects 0.000 claims description 9
- 238000005070 sampling Methods 0.000 claims description 4
- 238000000605 extraction Methods 0.000 claims description 2
- 230000006870 function Effects 0.000 description 11
- 238000004364 calculation method Methods 0.000 description 4
- 238000011156 evaluation Methods 0.000 description 4
- 230000003287 optical effect Effects 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 238000004891 communication Methods 0.000 description 3
- 238000001514 detection method Methods 0.000 description 3
- 238000003745 diagnosis Methods 0.000 description 3
- 238000005259 measurement Methods 0.000 description 3
- 230000001052 transient effect Effects 0.000 description 3
- 230000006872 improvement Effects 0.000 description 2
- 230000007774 longterm Effects 0.000 description 2
- 230000008447 perception Effects 0.000 description 2
- 238000000513 principal component analysis Methods 0.000 description 2
- 230000000644 propagated effect Effects 0.000 description 2
- 230000002441 reversible effect Effects 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 230000003750 conditioning effect Effects 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 230000005674 electromagnetic induction Effects 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 230000010355 oscillation Effects 0.000 description 1
- 238000012827 research and development Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
- 230000009897 systematic effect Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R35/00—Testing or calibrating of apparatus covered by the other groups of this subclass
- G01R35/02—Testing or calibrating of apparatus covered by the other groups of this subclass of auxiliary devices, e.g. of instrument transformers according to prescribed transformation ratio, phase angle, or wattage rating
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Mathematical Analysis (AREA)
- Theoretical Computer Science (AREA)
- Mathematical Physics (AREA)
- Pure & Applied Mathematics (AREA)
- Computational Mathematics (AREA)
- Mathematical Optimization (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Operations Research (AREA)
- Probability & Statistics with Applications (AREA)
- Bioinformatics & Computational Biology (AREA)
- Algebra (AREA)
- Life Sciences & Earth Sciences (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention relates to a current transformer error estimation method and a device based on a diffusion model, wherein the method comprises the following steps: acquiring operation data of a target current transformer, wherein the operation data comprises a secondary current value, a specific difference, a phase difference, a current oscillogram and an instantaneous power spectrum; extracting specific difference noise, phase difference noise and instantaneous power spectrum noise corresponding to the current oscillogram from the operation data, matching the specific difference noise, the phase difference noise and the instantaneous power spectrum noise, and forming combined noise distribution; constructing a data set based on the joint noise distribution data and the operation data and training a potential diffusion model according to the data set; and inputting the real-time secondary current value of the target current transformer into the trained potential diffusion model to obtain the error estimation value of the target current transformer. The invention estimates errors through various error noise distributions and diffusion models of the current transformer, thereby improving the accuracy and the adaptability of the error estimation of the transformer.
Description
Technical Field
The invention belongs to the technical field of power equipment detection, and particularly relates to a current transformer error estimation method and device based on a diffusion model.
Background
CT (Current transformer) is one of the most important high-voltage devices in power systems, and is widely used in relay protection, current measurement and power system analysis. The traditional current transformer is of an electromagnetic induction type, and has the main advantages of relatively stable performance, suitability for long-term operation and long-term operation experience.
At present, fault diagnosis or error estimation of electric energy metering devices at home and abroad is not systematic and comprehensive enough to diagnose common typical faults on the aspects of local diagnosis and manual analysis; the state evaluation of the electric energy metering device depends too much on expert experience, the related circuit has more parameters and the models are not uniform, so that the objectivity and the accuracy are insufficient, and the adaptability is poor; therefore, the method has great improvement space in the aspects of establishment of the transformer state evaluation and error index system and selection of the state evaluation method.
In addition, due to the difference of operating environments, regions and economic conditions of power equipment, the existing fault diagnosis and state evaluation research and development results cannot be popularized and applied in a large area; maintenance of most power equipment needs to assess or evaluate the running state of the current transformer in a conventional mode of 'manual on-site verification + terminal metering automation system monitoring' according to local conditions; and more secondary detection equipment (such as a transformer error calibrator, an electric energy meter error calibrator, a PT secondary voltage drop error calibrator and the like) is adopted for equipment calibration, fault recording, loop detection and the like. Therefore, a future-oriented adaptive current transformer error estimation system is urgently needed.
Disclosure of Invention
In order to improve accuracy and adaptability of transformer error estimation and reduce difficulty of error estimation, a first aspect of the invention provides a current transformer error estimation method based on a diffusion model, which comprises the following steps: acquiring operation data of a target current transformer, wherein the operation data comprises a secondary current value, a specific difference, a phase difference, a current oscillogram and an instantaneous power spectrum; respectively extracting specific difference noise, phase difference noise and instantaneous power spectrum noise corresponding to the current oscillogram from the operating data based on the error precision of the target current transformer; matching the specific difference noise, the phase difference noise and the instantaneous power spectrum noise based on an orthogonal matching tracking method, and forming combined noise distribution; constructing a data set based on the joint noise distribution data and the operating data and training a potential diffusion model according to the data set; and inputting the real-time secondary current value of the target current transformer into the trained potential diffusion model to obtain the error estimation value of the target current transformer.
In some embodiments of the present invention, the extracting, from the operation data, a specific difference noise, a phase difference noise, and an instantaneous power spectrum noise corresponding to a current waveform diagram, respectively, based on the error accuracy of the target current transformer, includes: and carrying out multi-order difference on the secondary current value, the specific difference, the phase difference, the current oscillogram and the instantaneous power spectrum based on the error precision of the target current transformer to obtain specific difference noise, phase difference noise and power spectrum noise corresponding to the current oscillogram.
Further, the matching the specific difference noise, the phase difference noise and the instantaneous power spectrum noise based on the orthogonal matching pursuit method, and forming a joint noise distribution includes: respectively carrying out discrete cosine transform on the specific difference noise, the phase difference noise and the instantaneous power spectrum noise to obtain a plurality of sparse matrixes; and matching the plurality of sparse matrixes by an orthogonal matching pursuit method according to a preset sampling rate and a preset frame length to form joint noise distribution.
In some embodiments of the invention, the diffusion model comprises: an encoder for encoding the operational data of the current transformer into a first multi-dimensional vector and estimating one or more noise distributions therefrom; a decoder for decoding the first multi-dimensional vector into a current transformer error by an attention mechanism and one or more noise profiles; a conditional embedding module to embed, by an attention mechanism, one or more noise profiles into a decoder.
Further, the conditional embedding module embeds, by the transformer, the one or more noise distributions into a second multi-dimensional vector.
In the above embodiment, the constructing a data set and training a potential diffusion model according to the data set based on the joint noise distribution data and the operation data includes: randomly extracting one distribution in the combined noise distribution and a current oscillogram corresponding to the distribution as a sample, and taking the error of the current transformer as a label; and iterating the potential diffusion model by using the rest distribution based on a Markov probability model until the loss function value of the potential diffusion model is stable and lower than a threshold value, and obtaining the trained potential diffusion model.
In a second aspect of the present invention, a current transformer error estimation apparatus based on a diffusion model is provided, including: the acquisition module is used for acquiring operation data of the target current transformer, wherein the operation data comprises a secondary current value, a specific difference, a phase difference, a current oscillogram and an instantaneous power spectrum; the matching module is used for extracting specific difference noise, phase difference noise and instantaneous power spectrum noise corresponding to the current oscillogram from the operating data respectively based on the error precision of a target current transformer; matching the specific difference noise, the phase difference noise and the instantaneous power spectrum noise based on an orthogonal matching tracking method, and forming combined noise distribution; a training module for constructing a data set based on the joint noise distribution data and the operating data and training a potential diffusion model according to the data set; and the estimation module is used for inputting the real-time secondary current value of the target current transformer into the trained potential diffusion model to obtain the error estimation value of the target current transformer.
In a third aspect of the present invention, there is provided an electronic apparatus comprising: one or more processors; a storage device, configured to store one or more programs, which when executed by the one or more processors, cause the one or more processors to implement the current transformer error estimation method based on diffusion model provided in the first aspect of the present invention.
In a fourth aspect of the present invention, a computer-readable medium is provided, on which a computer program is stored, wherein the computer program, when executed by a processor, implements the diffusion model-based current transformer error estimation method provided in the first aspect of the present invention.
The invention has the beneficial effects that:
according to the method, various error noises of the current transformer are extracted, and due to the sparsity of the noises, the sparse representation of the current transformer is constructed through compressed sensing, so that the input data volume of a diffusion model can be effectively reduced, and the training efficiency of the current transformer is improved; meanwhile, multiple error noise distributions can improve the interpretability and the robustness of the diffusion model through a Markov probability model; on the other hand, the potential diffusion model reserves the data of the current transformer, meanwhile, diversified noise characteristics are generated, and potential characteristics of the transformer are reserved.
Drawings
FIG. 1 is a schematic flow chart of a basic method for diffusion model based error estimation of a current transformer in some embodiments of the present invention;
FIG. 2 is a graph of current waveforms and their corresponding error trends in some embodiments of the invention;
FIG. 3 is a schematic diagram of a potential diffusion model architecture in some embodiments of the present invention;
FIG. 4 is a schematic diagram of a current transformer error estimation apparatus based on a diffusion model according to some embodiments of the present invention;
fig. 5 is a schematic structural diagram of an electronic device in some embodiments of the invention.
Detailed Description
The principles and features of this invention are described below in conjunction with the following drawings, which are set forth by way of illustration only and are not intended to limit the scope of the invention.
Referring to fig. 1 and 3, in a first aspect of the present invention, there is provided a current transformer error estimation method based on a diffusion model, including: s100, obtaining operation data of a target current transformer, wherein the operation data comprises a secondary current value, a specific difference, a phase difference, a current oscillogram and an instantaneous power spectrum; s200, respectively extracting specific difference noise, phase difference noise and instantaneous power spectrum noise corresponding to a current oscillogram from the operating data based on the error precision of a target current transformer; matching the specific difference noise, the phase difference noise and the instantaneous power spectrum noise based on an orthogonal matching tracking method, and forming combined noise distribution; s300, constructing a data set based on the combined noise distribution data and the operation data, and training a potential diffusion model according to the data set; and S400, inputting the real-time secondary current value of the target current transformer into the trained potential diffusion model to obtain the error estimation value of the target current transformer.
In step S100 of some embodiments of the present invention, operational data of the target current transformer is acquired, the operational data including a secondary current value, a specific difference, a phase difference, a current waveform diagram, and an instantaneous power spectrum. And acquiring the operating data of the current transformer through a sensor, a wave recorder or a sampling circuit arranged on the current transformer.
It is understood that the above operation data generally refers to historical operation data of the target current transformer in the training phase for the diffusion model; however, the diffusion model is used as a generation model, missing characteristics or data can be supplemented, namely when the input operation data is one or more of real-time secondary current value, specific difference, phase difference, current oscillogram and instantaneous power spectrum, the error of the current transformer estimated by the diffusion model is not influenced by the attention mechanism and the conditional probability, so that the adaptability of the model is improved.
The jjjg 1021-2007 power transformer standard specifies specific and angular difference thresholds for CT at rated current percentages of 1%, 5%, 20%, 100%, 120%. And defining a specific difference threshold value CME _ base and an angular difference threshold value CPE _ base under the rated current percentages of 1%, 5%, 20%, 100% and 120% of the current transformer, and acquiring a specific difference measurement value CME and an angular difference measurement value CPE under the rated current percentages of 1%, 5%, 20%, 100% and 120% of the current transformer by a current transformer calibrator. If the absolute value CME is larger than CME _ base, judging that the ratio difference of the current transformer is out of tolerance; if the absolute CPE is larger than the CPE _ base, the current transformer angular difference is judged to be out of tolerance.
Referring to fig. 2, in step S200 of some embodiments of the present invention, the extracting, from the operation data, a specific difference noise, a phase difference noise, and an instantaneous power spectrum noise corresponding to a current waveform diagram, respectively, based on the error accuracy of the target current transformer, includes: and carrying out multi-order difference on the secondary current value, the specific difference, the phase difference, the current oscillogram and the instantaneous power spectrum based on the error precision of the target current transformer to obtain specific difference noise, phase difference noise and power spectrum noise corresponding to the current oscillogram.
Specifically, the error precision (1%, 5%, 20%, 100%, 120%) is adopted to filter out the specific difference noise, the phase difference noise and the instantaneous power spectrum noise which are higher than the error precision, so that the consistency of the subsequent model estimation error can be maintained. And then solving the first-order difference or the multi-order difference of the plurality of operation data to obtain the corresponding noise.
Optionally, solving the noise of the plurality of operation data through the Allan variance, that is: the method comprises the following steps of accurately extracting the fluctuation condition of an error sequence on a certain specified time scale, wherein the specific calculation steps are as follows: 1. partitioning the whole error sequence according to the length (for example, 1 minute) of a preset time scale (frame length); 2. averaging each block; 3. calculating the difference of the average values of the adjacent blocks; 4. and counting all the differences to obtain a mean square value, and multiplying by 1/2.
Further, the matching the specific difference noise, the phase difference noise and the instantaneous power spectrum noise based on the orthogonal matching pursuit method, and forming a joint noise distribution includes: respectively carrying out discrete cosine transform on the specific difference noise, the phase difference noise and the instantaneous power spectrum noise to obtain a plurality of sparse matrixes; and matching the plurality of sparse matrixes by an orthogonal matching pursuit method according to a preset sampling rate and a preset frame length to form joint noise distribution.
Specifically, a general DCT (discrete fourier transform) is defined as follows:
wherein,x(i) Is the ith term in the input time domain sequence (waveform or power spectrum),y( j) Is the j-th term of the output frequency domain sequenceC(j) Is defined as follows:
optionally, the vector constructed by the one or more noises is subjected to feature decomposition by using KLT (Karhunen-Leece Transform) or Principal Component Analysis (PCA) to obtain a sparse vector. Meanwhile, noise signals of the circuit usually include thermal noise of components and parts, and disturbance such as sag, rise, short-time interruption, pulse transient, oscillation transient, harmonic wave and flicker of the circuit where the circuit transformer is located, and based on 7 kinds of single disturbance and 40 kinds of mixed disturbance, the sparse matrixes are matched through an orthogonal matching tracking method to form combined noise distribution.
Referring to fig. 3, in step S300 or step S400 of some embodiments of the invention, the diffusion model includes: an encoder for encoding the operational data of the current transformer into a first multi-dimensional vector and estimating one or more noise distributions therefrom; a decoder for decoding the first multi-dimensional vector into a current transformer error by an attention mechanism and one or more noise profiles; more specifically, an Auto Encoder (AE) architecture is one that captures perceptual compression. The encoder in AE projects the high dimensional data into a Latent space (also called hidden space) and the decoder recovers the original data from the Latent space (patent space).
A conditional embedding module (conditioning) for embedding one or more noise profiles into the decoder by means of attention. Specifically, the diffusion model in the present embodiment is a prior-dependent conditional model (probability distribution). In this embodiment, the apriori is typically one or more of a secondary current value, a specific difference, a phase difference, a current waveform map, and an instantaneous power spectrum that have been verified.
To obtain a potential representation (signature) of this situation, a transformer (e.g., CLIP) is used that embeds the current transformer data signatures into the potential vectorIn (1). The resulting loss function thus depends not only on the latent space of the original current transformer data, but also on the latent embedding of the conditions.
Without loss of generality, in fig. 3:xwhich is indicative of the true error,is an error estimation value;Zfor a first multi-dimensional vector containing a plurality of types of noise,trepresenting the number of frames or rounds of training,Zta noise vector representing the prediction is determined,a multidimensional noise vector (second multidimensional vector) representing the posterior.Q、K、VA weight vector representing the attention mechanism,representing an encoder.
In the training process, the potential diffusion model comprises two stages, which respectively correspond to the perception loss and the diffusion loss. Wherein, the perception loss of the first stage Diffusion: the autoencoder captures the perceptual structure of the data by projecting the data into a latent space. The loss function ensures that the reconstruction is confined within the data manifold and reduces ambiguities that may occur when using data space losses (e.g., L1/L2 losses).
Where the diffusion loss function or objective function (perceptual loss) is expressed as:
In the second stage of learning (Denoising), the generation method (decoding) must be able to capture the characteristic information present in the data (noise, time domain information, frequency domain information and power spectrum information). The combination of the generalization capability of the Transformer and the detail retention capability of the diffusion model integrates the advantages of the two, provides the capability of generating fine-grained noise or errors, and simultaneously retains the time domain information, the frequency domain information and the power spectrum information which are implicit in the current waveform.
The backbone of the LDM is a U-Net self-encoder with sparse connections, providing a cross-attention mechanism. The Transformer network encodes the above noise, time domain information, frequency domain information, and power spectrum information as potential embeddings and then maps to the middle layer of the U-Net through the cross attention layer. This cross-attention layer implements the attention mechanism, and Q, K, and V are learnable projection matrices:
the diffusion model learns the data distribution by gradually removing noise in the normal distribution variables. DMs employ reverse Markov chains of length T. This also means that the DMs can be modeled as a series of T denoised auto-encoders with time steps T =1, \ 8230;, T. This is represented byAnd (4) showing. It should be noted that the loss function depends on implicit vectors rather than vector space.
Thus, the diffusion loss of the potential diffusion model is expressed as:
in step S300 of the above embodiment, the constructing a data set and training a potential diffusion model according to the data set based on the joint noise distribution data and the operation data includes: randomly extracting one distribution in the combined noise distribution and a current oscillogram corresponding to the distribution as a sample, and taking the error of the current transformer as a label; and iterating the residual distribution to the potential diffusion model based on a Markov probability model until the loss function value of the potential diffusion model is stable and is lower than a threshold value, so as to obtain the trained potential diffusion model.
Optionally, the characteristic data or the joint noise distribution includes, but is not limited to, a secondary current value, a specific difference, a phase difference, a current waveform diagram, a voltage waveform diagram, an instantaneous power spectrum, and other transient or steady characteristics (time constant, hysteresis coefficient) generated during the operation of the current transformer. The error calculation between the distributions can adopt cross entropy, wasserstein distance and other joint probability distribution distances for calculation.
It can be understood that through sparse representation (compressed sensing) of noise distribution of various components of the current transformer error, the accuracy and robustness of the diffusion model can be improved, and meanwhile, the data amount and the calculation amount of the model in the noise generation or reasoning process can be effectively reduced.
Example 2
Referring to fig. 4, in a second aspect of the present invention, there is provided a current transformer error estimation apparatus 1 based on a diffusion model, including: the acquisition module 11 is configured to acquire operation data of a target current transformer, where the operation data includes a secondary current value, a specific difference, a phase difference, a current oscillogram, and an instantaneous power spectrum; the matching module 12 is configured to extract, from the operating data, a specific difference noise, a phase difference noise, and an instantaneous power spectrum noise corresponding to the current oscillogram, respectively, based on the error accuracy of the target current transformer; matching the specific difference noise, the phase difference noise and the instantaneous power spectrum noise based on an orthogonal matching tracking method, and forming combined noise distribution; a training module 13, configured to construct a data set based on the joint noise distribution data and the operation data, and train a potential diffusion model according to the data set; and the estimation module 14 is configured to input the real-time secondary current value of the target current transformer to the trained potential diffusion model, so as to obtain an error estimation value of the target current transformer.
In some embodiments of the invention, the matching module 12 comprises: and the extraction unit is used for carrying out multi-order difference on the secondary current value, the specific difference, the phase difference, the current oscillogram and the instantaneous power spectrum based on the error precision of the target current transformer to obtain specific difference noise, phase difference noise and power spectrum noise corresponding to the current oscillogram. And the matching unit is used for matching the specific difference noise, the phase difference noise and the instantaneous power spectrum noise based on an orthogonal matching tracking method and forming combined noise distribution.
Example 3
Referring to fig. 5, in a third aspect of the present invention, there is provided an electronic apparatus comprising: one or more processors; a storage device for storing one or more programs which, when executed by the one or more processors, cause the one or more processors to implement the diffusion model-based current transformer error estimation method of the present invention in the first aspect.
The electronic device 500 may include a processing means (e.g., central processing unit, graphics processor, etc.) 501 that may perform various appropriate actions and processes in accordance with a program stored in a Read Only Memory (ROM) 502 or a program loaded from a storage means 508 into a Random Access Memory (RAM) 503. In the RAM 503, various programs and data necessary for the operation of the electronic apparatus 500 are also stored. The processing device 501, the ROM 502, and the RAM 503 are connected to each other through a bus 504. An input/output (I/O) interface 505 is also connected to bus 504.
The following devices may be connected to the I/O interface 505 in general: input devices 506 including, for example, a touch screen, touch pad, keyboard, mouse, camera, microphone, accelerometer, gyroscope, etc.; output devices 507 including, for example, a Liquid Crystal Display (LCD), speakers, vibrators, and the like; a storage device 508 including, for example, a hard disk; and a communication device 509. The communication means 509 may allow the electronic device 500 to communicate with other devices wirelessly or by wire to exchange data. While fig. 5 illustrates an electronic device 500 having various means, it is to be understood that not all illustrated means are required to be implemented or provided. More or fewer devices may alternatively be implemented or provided. Each block shown in fig. 5 may represent one device or may represent multiple devices as desired.
In particular, according to an embodiment of the present disclosure, the processes described above with reference to the flowcharts may be implemented as computer software programs. For example, embodiments of the present disclosure include a computer program product comprising a computer program embodied on a computer readable medium, the computer program comprising program code for performing the method illustrated in the flow chart. In such an embodiment, the computer program may be downloaded and installed from a network via the communication means 509, or installed from the storage means 508, or installed from the ROM 502. The computer program, when executed by the processing device 501, performs the above-described functions defined in the methods of embodiments of the present disclosure. It should be noted that the computer readable medium described in the embodiments of the present disclosure may be a computer readable signal medium or a computer readable storage medium or any combination of the two. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples of the computer readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, 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 disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In embodiments of the disclosure, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In embodiments of the present disclosure, however, a computer readable signal medium may comprise a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: electrical wires, optical cables, RF (radio frequency), etc., or any suitable combination of the foregoing.
The computer readable medium may be embodied in the electronic device; or may exist separately without being assembled into the electronic device. The computer readable medium carries one or more computer programs which, when executed by the electronic device, cause the electronic device to:
computer program code for carrying out operations for embodiments of the present disclosure may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, smalltalk, C + +, python, and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider).
The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (10)
1. A current transformer error estimation method based on a diffusion model is characterized by comprising the following steps:
acquiring operation data of a target current transformer, wherein the operation data comprises a secondary current value, a specific difference, a phase difference, a current oscillogram and an instantaneous power spectrum;
respectively extracting specific difference noise, phase difference noise and instantaneous power spectrum noise corresponding to the current oscillogram from the operation data based on the error precision of the target current transformer; matching the specific difference noise, the phase difference noise and the instantaneous power spectrum noise based on an orthogonal matching tracking method, and forming combined noise distribution;
constructing a data set based on the joint noise distribution data and the operation data and training a potential diffusion model according to the data set;
and inputting the real-time secondary current value of the target current transformer into the trained potential diffusion model to obtain the error estimation value of the target current transformer.
2. The diffusion model-based current transformer error estimation method of claim 1, wherein the extracting, based on the error accuracy of the target current transformer, a specific difference noise, a phase difference noise, and an instantaneous power spectrum noise corresponding to a current waveform diagram from the operational data, respectively, comprises:
and carrying out multi-order difference on the secondary current value, the specific difference, the phase difference, the current oscillogram and the instantaneous power spectrum based on the error precision of the target current transformer to obtain specific difference noise, phase difference noise and power spectrum noise corresponding to the current oscillogram.
3. The diffusion model-based current transformer error estimation method according to claim 2, wherein the matching the specific difference noise, the phase difference noise and the instantaneous power spectrum noise and constructing a joint noise distribution based on the orthogonal matching pursuit method comprises:
respectively carrying out discrete cosine transform on the specific difference noise, the phase difference noise and the instantaneous power spectrum noise to obtain a plurality of sparse matrixes;
and matching the sparse matrixes by an orthogonal matching pursuit method according to a preset sampling rate and a preset frame length to form combined noise distribution.
4. The diffusion model-based current transformer error estimation method of claim 1, wherein the diffusion model comprises:
an encoder for encoding the operational data of the current transformer into a first multi-dimensional vector and estimating one or more noise distributions therefrom;
a decoder for decoding the first multi-dimensional vector into a current transformer error by an attention mechanism and one or more noise profiles;
a conditional embedding module to embed, by a mechanism of attention, one or more noise profiles into the decoder.
5. The diffusion model-based current transformer error estimation method of claim 4, wherein the conditional embedding module embeds one or more noise distributions into a second multi-dimensional vector through a transformer.
6. The diffusion model-based current transformer error estimation method of any one of claims 1 to 5, wherein the constructing a data set and training a potential diffusion model based thereon based on the joint noise distribution data and the operational data comprises:
randomly extracting one distribution in the combined noise distribution and a current oscillogram corresponding to the distribution as a sample, and taking the error of the current transformer as a label;
and iterating the potential diffusion model by using the rest distribution based on a Markov probability model until the loss function value of the potential diffusion model is stable and is lower than a threshold value, so as to obtain the trained potential diffusion model.
7. A current transformer error estimation device based on a diffusion model is characterized by comprising:
the acquisition module is used for acquiring operation data of the target current transformer, wherein the operation data comprises a secondary current value, a specific difference, a phase difference, a current oscillogram and an instantaneous power spectrum;
the matching module is used for extracting specific difference noise, phase difference noise and instantaneous power spectrum noise corresponding to the current oscillogram from the operating data respectively based on the error precision of a target current transformer; matching the specific difference noise, the phase difference noise and the instantaneous power spectrum noise based on an orthogonal matching tracking method, and forming combined noise distribution;
a training module for constructing a data set based on the joint noise distribution data and the operating data and training a potential diffusion model according to the data set;
and the estimation module is used for inputting the real-time secondary current value of the target current transformer into the trained potential diffusion model to obtain the error estimation value of the target current transformer.
8. The diffusion model-based current transformer error estimation apparatus of claim 7, wherein the matching module comprises:
the extraction unit is used for carrying out multi-order difference on the secondary current value, the specific difference, the phase difference, the current oscillogram and the instantaneous power spectrum based on the error precision of the target current transformer to obtain specific difference noise, phase difference noise and power spectrum noise corresponding to the current oscillogram;
and the matching unit is used for matching the specific difference noise, the phase difference noise and the instantaneous power spectrum noise based on an orthogonal matching tracking method and forming combined noise distribution.
9. An electronic device, comprising: one or more processors; a storage device to store one or more programs that, when executed by the one or more processors, cause the one or more processors to implement the diffusion model based current transformer error estimation method of any one of claims 1 to 6.
10. A computer-readable medium, characterized in that a computer program is stored thereon, wherein the computer program, when being executed by a processor, realizes the diffusion model based current transformer error estimation method according to any one of claims 1 to 6.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211478921.XA CN115542230B (en) | 2022-11-24 | 2022-11-24 | Current transformer error estimation method and device based on diffusion model |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211478921.XA CN115542230B (en) | 2022-11-24 | 2022-11-24 | Current transformer error estimation method and device based on diffusion model |
Publications (2)
Publication Number | Publication Date |
---|---|
CN115542230A CN115542230A (en) | 2022-12-30 |
CN115542230B true CN115542230B (en) | 2023-03-24 |
Family
ID=84719681
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202211478921.XA Active CN115542230B (en) | 2022-11-24 | 2022-11-24 | Current transformer error estimation method and device based on diffusion model |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115542230B (en) |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5461306A (en) * | 1993-12-28 | 1995-10-24 | Schlumberger Industries, Inc. | DC current detection system for a current transformer |
CN101644752A (en) * | 2009-08-14 | 2010-02-10 | 河南电力试验研究院 | Online accuracy detection method and system of current transformer |
JP2013053855A (en) * | 2011-08-31 | 2013-03-21 | National Institute Of Advanced Industrial & Technology | Error estimation device for electronic current transformer |
CN106054102A (en) * | 2016-08-11 | 2016-10-26 | 国网浙江省电力公司电力科学研究院 | Current-transformer harmonic wave error measurement system |
KR101908369B1 (en) * | 2017-06-07 | 2018-10-17 | 한국전력공사 | Apparatus and computer readable recording medium for determining ratio error of metering outfit |
CN110095744A (en) * | 2019-04-04 | 2019-08-06 | 国网江苏省电力有限公司电力科学研究院 | A kind of electronic mutual inductor error prediction method |
CN110632545A (en) * | 2019-09-25 | 2019-12-31 | 国网江苏省电力有限公司电力科学研究院 | Regression neural network-based error risk assessment method and device for electronic transformer |
-
2022
- 2022-11-24 CN CN202211478921.XA patent/CN115542230B/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5461306A (en) * | 1993-12-28 | 1995-10-24 | Schlumberger Industries, Inc. | DC current detection system for a current transformer |
CN101644752A (en) * | 2009-08-14 | 2010-02-10 | 河南电力试验研究院 | Online accuracy detection method and system of current transformer |
JP2013053855A (en) * | 2011-08-31 | 2013-03-21 | National Institute Of Advanced Industrial & Technology | Error estimation device for electronic current transformer |
CN106054102A (en) * | 2016-08-11 | 2016-10-26 | 国网浙江省电力公司电力科学研究院 | Current-transformer harmonic wave error measurement system |
KR101908369B1 (en) * | 2017-06-07 | 2018-10-17 | 한국전력공사 | Apparatus and computer readable recording medium for determining ratio error of metering outfit |
CN110095744A (en) * | 2019-04-04 | 2019-08-06 | 国网江苏省电力有限公司电力科学研究院 | A kind of electronic mutual inductor error prediction method |
CN110632545A (en) * | 2019-09-25 | 2019-12-31 | 国网江苏省电力有限公司电力科学研究院 | Regression neural network-based error risk assessment method and device for electronic transformer |
Non-Patent Citations (2)
Title |
---|
基于MSP430的智能型电流互感器校验仪的设计与实现;夏阳;《中国优秀硕士学位论文全文数据库 (工程科技Ⅱ辑)》;20130715;全文 * |
能源互联网电能数据误差成因分析研究;韩云飞;《中国优秀硕士学位论文全文数据库(工程科技Ⅱ辑)》;20220315;全文 * |
Also Published As
Publication number | Publication date |
---|---|
CN115542230A (en) | 2022-12-30 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10909419B2 (en) | Abnormality detection device, learning device, abnormality detection method, and learning method | |
CN103488881B (en) | Equipment residual service life prediction method under the condition of uncertain degradation measured data | |
CN113777552B (en) | Method and system for predicting metering error of capacitive voltage transformer | |
CN108549908B (en) | Chemical process fault detection method based on multi-sampling probability kernel principal component model | |
CN112668238B (en) | Rainfall processing method, rainfall processing device, rainfall processing equipment and storage medium | |
Jiang et al. | Literature review of power disaggregation | |
CN116320042A (en) | Internet of things terminal monitoring control system for edge calculation | |
CN112148557A (en) | Method for predicting performance index in real time, computer equipment and storage medium | |
CN110852906B (en) | Method and system for identifying electricity stealing suspicion based on high-dimensional random matrix | |
CN113723861A (en) | Abnormal electricity consumption behavior detection method and device, computer equipment and storage medium | |
CN115392301A (en) | Converter transformer state identification method, converter transformer state identification device, converter transformer state identification equipment, converter transformer state identification medium and program product | |
CN113284001A (en) | Power consumption prediction method and device, computer equipment and storage medium | |
CN116662771A (en) | Method and system for self-adaptive PCA error quantitative evaluation of capacitive voltage transformer | |
CN114881157A (en) | Method, device and equipment for detecting working state of converter valve and storage medium | |
CN117290788B (en) | Power distribution network fault identification method and system based on improved wavelet transformation algorithm | |
CN115542230B (en) | Current transformer error estimation method and device based on diffusion model | |
CN116777452B (en) | Prepayment system and method for intelligent ammeter | |
CN117668427A (en) | Temperature monitoring and predicting method and equipment for electricity consumption information system based on data analysis | |
CN117290685A (en) | Power plant power equipment expert diagnosis system and method based on historical data | |
CN114897032B (en) | Current transformer fault diagnosis method and device based on width learning | |
CN115144807A (en) | Differential noise filtering and current-carrying grading current transformer online evaluation method and device | |
CN110969238A (en) | Method and device for calibrating electricity consumption data | |
CN117526318B (en) | Power distribution network state evaluation method and device, electronic equipment and storage medium | |
CN118094205B (en) | Intelligent tunnel structure health monitoring system based on deep learning | |
CN116125922B (en) | Complex industrial process monitoring method and system based on parallel dictionary learning |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |