CN116609612A - Multi-harmonic source identification method and system for power distribution network - Google Patents
Multi-harmonic source identification method and system for power distribution network Download PDFInfo
- Publication number
- CN116609612A CN116609612A CN202310573380.7A CN202310573380A CN116609612A CN 116609612 A CN116609612 A CN 116609612A CN 202310573380 A CN202310573380 A CN 202310573380A CN 116609612 A CN116609612 A CN 116609612A
- Authority
- CN
- China
- Prior art keywords
- harmonic
- mutual information
- value
- power distribution
- distribution network
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 52
- 238000013136 deep learning model Methods 0.000 claims abstract description 21
- 238000013528 artificial neural network Methods 0.000 claims description 21
- 230000006870 function Effects 0.000 claims description 19
- 230000008569 process Effects 0.000 claims description 15
- 239000011159 matrix material Substances 0.000 claims description 9
- 238000012937 correction Methods 0.000 claims description 8
- 238000007781 pre-processing Methods 0.000 claims description 6
- 230000002087 whitening effect Effects 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 5
- 238000005457 optimization Methods 0.000 claims description 4
- 230000009466 transformation Effects 0.000 claims description 4
- 230000009977 dual effect Effects 0.000 claims description 3
- 238000005070 sampling Methods 0.000 claims description 3
- 230000001537 neural effect Effects 0.000 claims description 2
- 238000012545 processing Methods 0.000 description 4
- 238000000926 separation method Methods 0.000 description 4
- 238000012549 training Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000010248 power generation Methods 0.000 description 1
- 238000000844 transformation Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/08—Locating faults in cables, transmission lines, or networks
- G01R31/081—Locating faults in cables, transmission lines, or networks according to type of conductors
- G01R31/086—Locating faults in cables, transmission lines, or networks according to type of conductors in power transmission or distribution networks, i.e. with interconnected conductors
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/08—Locating faults in cables, transmission lines, or networks
- G01R31/088—Aspects of digital computing
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E40/00—Technologies for an efficient electrical power generation, transmission or distribution
- Y02E40/40—Arrangements for reducing harmonics
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Theoretical Computer Science (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention provides a multi-harmonic source identification method of a power distribution network, which comprises the steps of obtaining harmonic voltage signals generated by superposition of a plurality of harmonic sources on PCC nodes, and measuring harmonic voltage values; separating a fast variation component and a slow variation component in the harmonic voltage signal, and estimating a harmonic current value injected by the PCC node based on the fast variation component; the obtained harmonic current value and harmonic voltage value are led into a pre-trained mutual information deep learning model, so as to obtain the mutual information value between the harmonic current emitted by each harmonic source and the harmonic voltage of the PCC node, and the main harmonic source is identified according to each obtained mutual information value. By implementing the method, the problem that the existing method is difficult to effectively identify and position the multi-harmonic source in the complex power distribution network can be solved.
Description
Technical Field
The invention relates to the technical field of power system detection, in particular to a method and a system for identifying multiple harmonic sources of a power distribution network.
Background
In recent years, the construction of novel power systems is continuously advanced, challenges facing the power grid are more complex and multiple, and particularly, the 'double high' characteristics lead to superposition of power quality problems and more complex characteristics, so that higher requirements are put on the power quality. Along with the high-density access of the distributed power generation and power electronic equipment to the power grid, the number of harmonic sources is increased, the running state is changeable, and the harmonic pollution is increased. Therefore, the harmonic source positioning and identification have important roles in defining the harmonic pollution source, and are the precondition of harmonic responsibility division and economic disputes solving.
The traditional harmonic source identification is mainly based on a mechanism model from the aspect of electric science, and mainly adopts an equivalent circuit model, a method based on harmonic state estimation and harmonic impedance, and the like. However, these methods suffer from a number of factors, and it is difficult to effectively identify and locate multiple harmonic sources in a complex distribution network.
Therefore, a new method for identifying multiple harmonic sources of a power distribution network is needed, and the problem that the existing method is difficult to effectively identify and position multiple harmonic sources in a complex power distribution network is solved.
Disclosure of Invention
The technical problem to be solved by the embodiment of the invention is to provide a method and a system for identifying multiple harmonic sources of a power distribution network, which can solve the problem that the existing method is difficult to effectively identify and position the multiple harmonic sources in the complex power distribution network.
In order to solve the technical problems, the embodiment of the invention provides a method for identifying multiple harmonic sources of a power distribution network, which comprises the following steps:
acquiring harmonic voltage signals generated by superposition of a plurality of harmonic sources on PCC nodes, and measuring harmonic voltage values;
separating a fast variation component and a slow variation component in the harmonic voltage signal, and estimating a harmonic current value injected by the PCC node based on the fast variation component;
and importing the obtained harmonic current value and the harmonic voltage value into a pre-trained mutual information deep learning model to obtain a mutual information value between the harmonic current emitted by each harmonic source and the harmonic voltage of the PCC node, and identifying a main harmonic source according to each obtained mutual information value.
Wherein the harmonic voltage signal is passed through a filter to separate the fast-varying component and the slow-varying component.
Wherein, the harmonic current value injected by the PCC node is realized by a FastICA algorithm; wherein,
the FastICA algorithm includes the steps of: and preprocessing data, and establishing an objective function for optimizing.
Wherein the data preprocessing comprises a decentralization processing and a whitening processing; wherein,
the de-centering process refers to subtracting the average value of all the sampled signals to obtain a group of original data with zero average value: wherein ,X(ti ) Is a sampled signal;
the whitening process is a process of decorrelating the sampled data such that the observed signal X has a unit variance: x=ed -1/2 E TX； wherein ,consists of n eigenvalues; d, d i The corresponding feature vector is c i ，E＝[c 1 ,c 2 ,...c n ]The method comprises the steps of carrying out a first treatment on the surface of the The observation signal x=as, a is a mixing matrix, and S is a harmonic current source injected into the PCC node.
Wherein the process of establishing the objective function for optimizing is to find a direction so as to output w T X(y＝w T X) maximum non-gaussian; wherein,
non-gaussian properties are measured by an approximation of negative entropy: n (N) g (Y)＝{E[g(Y)]-E[g(Y Gauss )]} 2 I.e. find J G (w)＝[E{G(w T X)}] 2 A maximum value; where W is an m-dimensional variable representing one row of the unmixed matrix W;
the objective function is defined as:according to the Kunhn-Tucker condition, the problem of unconstrained optimization is translated, so that the objective function is transformed into: f (w) =e [ G (w) T X)]+C(||w|| 2 -1); the optimal solution of the objective function is obtained through a Newton iteration method: />
The mutual information deep learning model is constructed based on a neural network; wherein,
the neural network comprises an input layer, a plurality of hidden layers and an output layer; the input layer receives samples of the harmonic current value Y and the harmonic voltage value X as input, and the hidden layer obtains a result Z of the output layer after a series of nonlinear transformation.
The neural network is obtained by training by executing the following steps:
7.1 initializing neural network parameters;
7.2 extracting b miniband samples from the joint distribution; wherein the b miniband samples extracted in the joint distribution are denoted (x) (1) ,y (1) ),...,(x (b) ,y (b) ) Obeying joint probability distribution
7.3 extracting b samples from the Y edge distribution; wherein the b samples extracted from the Y edge distribution are recorded asObeying the joint probability distribution->
7.4, evaluating the mutual information lower bound; the mutual information lower bound calculation formula is as follows:
7.5 correcting the deviation correction gradient by EMA; wherein the EMA correction bias correction gradient is expressed as:
7.6, updating parameters of the neural network; wherein the parameters of the updated neural network are
7.7 repeating the steps 7.1 to 7.6 until the convergence condition is reached.
Wherein, the mutual information lower bound between X and Y is calculated by utilizing KL divergence calculated by using dual form:
where T is a function defined on X Y that all makes two expectations finite;and I (X; Y) is not less than I θ (X,Y)；/>Is the amount of neural information and is expressed by referring to the gradient formulaDescent to maximize;
wherein, the mutual information deep learning model can be represented by a formulaTo express, wherein->The empirical distribution of distribution P given n independent samplings is represented.
The embodiment of the invention also provides a system for identifying the multiple harmonic sources of the power distribution network, which comprises the following steps:
the harmonic voltage acquisition unit is used for acquiring harmonic voltage signals generated by superposition of a plurality of harmonic sources on the PCC node and measuring harmonic voltage values;
the harmonic current estimation unit is used for separating a fast variation component and a slow variation component in the harmonic voltage signal and estimating a harmonic current value injected by the PCC node based on the fast variation component;
the main harmonic source identification unit is used for leading the obtained harmonic current value and the harmonic voltage value into a pre-trained mutual information deep learning model to obtain a mutual information value between the harmonic current emitted by each harmonic source and the harmonic voltage of the PCC node, and identifying a main harmonic source according to each obtained mutual information value.
The embodiment of the invention has the following beneficial effects:
1. according to the invention, only the harmonic voltage signal of the PCC node is required to be obtained, the harmonic current is estimated by using the FastICA algorithm, and the mutual information is estimated through the mutual information deep learning model to identify the main harmonic source, so that the operation is simple, the position of the harmonic current source can be identified under the condition that the system network parameters are unknown, and the problem that the multi-harmonic source in the complex power distribution network is difficult to effectively identify and position by the existing method is solved;
2. the FastICA algorithm in the invention estimates the harmonic current, so that the convergence speed is high, the separation effect is good, the iteration is stable, the separation of non-Gaussian independent components can be carried out, and the accuracy and the reliability of the harmonic current estimation can be improved;
3. the mutual information deep learning model in the invention has simple calculation, can be better adapted to complex nonlinear relations in the power grid, and has stronger universality and reliability.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions of the prior art, the drawings which are required in the description of the embodiments or the prior art will be briefly described, it being obvious that the drawings in the description below are only some embodiments of the invention, and that it is within the scope of the invention to one skilled in the art to obtain other drawings from these drawings without inventive faculty.
Fig. 1 is a flowchart of a method for identifying multiple harmonic sources of a power distribution network according to an embodiment of the present invention;
fig. 2 is a flowchart of neural network training in a multi-harmonic source identification method of a power distribution network according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram of a multi-harmonic source identification system of a power distribution network according to an embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings, for the purpose of making the objects, technical solutions and advantages of the present invention more apparent.
The inventors have found that when a nonlinear load is connected to a distribution network, the harmonic currents emitted by different harmonic sources are not completely random and independent of the harmonic voltages at the PCC nodes, but that there is a relationship which can be characterized by mutual information (Mutual Information, MI). Therefore, in order to effectively improve accuracy of mutual information estimation, the inventor proposes a mutual information deep learning model for multi-harmonic source identification of a power distribution network, and the complex nonlinear relation in the power distribution network can be better adapted to through the deep learning model, so that the method has stronger universality and reliability and is simple to calculate.
As shown in fig. 1, in an embodiment of the present invention, a method for identifying multiple harmonic sources of a power distribution network is provided, where the method includes the following steps:
s1, acquiring harmonic voltage signals generated by superposition of a plurality of harmonic sources on a PCC node, and measuring a harmonic voltage value;
s2, separating a fast variation component and a slow variation component in the harmonic voltage signal, and estimating a harmonic current value injected by the PCC node based on the fast variation component;
and S3, importing the obtained harmonic current value and the obtained harmonic voltage value into a pre-trained mutual information deep learning model to obtain a mutual information value between the harmonic current emitted by each harmonic source and the harmonic voltage of the PCC node, and identifying a main harmonic source according to each obtained mutual information value.
The specific process is that before step S1, a mutual information deep learning model based on a neural network is constructed and trained, and the specific steps are as follows:
firstly, determining the structure of a neural network, wherein the neural network comprises an input layer, a plurality of hidden layers and an output layer; the input layer receives a sample of the harmonic current value Y and the harmonic voltage value X as input, and the hidden layer obtains a result Z of the output layer after a series of nonlinear transformations.
Next, as shown in fig. 2, training the neural network is specifically:
(1) Initializing neural network parameters;
(2) Extracting b miniband samples from the joint distribution; wherein the b miniband samples extracted in the joint distribution are denoted (x) (1) ,y (1) ),...,(x (b) ,y (b) ) Obeying joint probability distribution
(3) Extracting b samples from the Y edge distribution; wherein b samples extracted from the Y-edge distribution are denoted asObeying the joint probability distribution->
(4) Evaluating a mutual information lower bound; the mutual information lower bound calculation formula is as follows:
(5) Correcting the deviation correction gradient by EMA; wherein the EMA corrected deviation correction gradient is expressed as:
(6) Updating parameters of the neural network; wherein the parameters of the updated neural network are
(7) And (3) repeating the steps (1) to (6) until convergence conditions are reached, and obtaining a trained mutual information deep learning model.
It should be noted that in the neural network, the lower bound of mutual information between X and Y is calculated using KL divergence calculated using the dual form:
where T is a function defined on X Y that all makes two expectations finite;and I (X; Y) is not less than I θ (X, Y); i (X; Y) =H (X) -H (X|Y), where the mutual information between X and Y is equivalent to the KL divergence of the product of the joint probability distribution and the edge probability distribution is +.>H is information entropy, H (X|Y) is conditional information entropy,
recorded as neuro-informative, can be maximized with gradient descent, where the gradient formula is +.>
At this time, the mutual information deep learning model can be represented by the formulaTo express, wherein->The empirical distribution of distribution P given n independent samplings is represented.
In step S1, harmonic voltage signals generated by a plurality of harmonic sources superimposed on the PCC node are directly obtained, and only the harmonic voltage value X needs to be measured.
In step S2, first, a filter is used to separate out a fast-varying component and a slow-varying component in the harmonic voltage signal.
Secondly, based on the fast-varying component, the harmonic current value Y injected by the PCC node is estimated through a FastICA algorithm, specifically as follows:
(1) Setting an observation signal X to represent the harmonic voltage of the PCC node; the hybrid matrix a represents an admittance matrix Z; the source signal S represents an injected harmonic current source, and the estimation matrix Y represents an estimated harmonic current value;
(2) Estimating a Y value through a FastICA algorithm to enable the Y value to approach a harmonic current source infinitely, wherein the method specifically comprises the steps of data preprocessing and target function establishment for optimizing;
the data preprocessing comprises a decentralization processing and a whitening processing; the decentralization process refers to subtracting the average value of all the sampled signals to obtain a set of original data with zero average value: wherein ,X(ti ) Is a sampled signal; the whitening process is a process of decorrelating the sampled data such that the observed signal X has a unit variance: x=ed -1/2 E T X is a group; wherein (1)>Consists of n eigenvalues; d, d i The corresponding feature vector is c i ，E＝[c 1 ,c 2 ,...c n ]The method comprises the steps of carrying out a first treatment on the surface of the Observing signal X=AS, A is a mixed matrix, S is a harmonic current source injected by PCC nodes;
wherein, the process of establishing the objective function for optimizing is to find a direction to output w T X(y＝w T X) maximum non-gaussian; at this point, the non-gaussian property is measured by an approximation of the negative entropy: n (N) g (Y)＝{E[g(Y)]-E[g(Y Gauss )]} 2 I.e. find J G (w)＝[E{G(w T X)}] 2 A maximum value; where W is an m-dimensional variable representing one row of the unmixed matrix W;
defining an objective function as:according to the Kunhn-turner condition, the transformation is an unconstrained optimization problem, so that the objective function is transformed into F (w) =e [ G (w T X)]+C(||w|| 2 -1); the optimal solution of the objective function is obtained through a Newton iteration method: />
In step S3, first, a harmonic current value Y and a harmonic voltage value X are imported into a trained mutual information deep learning model to obtain a mutual information value between a harmonic current emitted by each harmonic source and a PCC node harmonic voltage; next, the main harmonic source is identified based on the obtained mutual information values. In one example, a harmonic current corresponding to the maximum mutual information value is found, and a corresponding emitted harmonic source, namely a main harmonic source, is determined based on the found harmonic current.
As shown in fig. 3, in an embodiment of the present invention, a power distribution network multi-harmonic source identification system is provided, including:
a harmonic voltage obtaining unit 110, configured to obtain harmonic voltage signals generated by superimposing a plurality of harmonic sources on the PCC node, and measure a harmonic voltage value;
a harmonic current estimation unit 120, configured to separate a fast-varying component and a slow-varying component in the harmonic voltage signal, and estimate a harmonic current value injected into the PCC node based on the fast-varying component;
the main harmonic source identification unit 130 is configured to introduce the obtained harmonic current value and the obtained harmonic voltage value into a pre-trained mutual information deep learning model, obtain a mutual information value between the harmonic current emitted by each harmonic source and the PCC node harmonic voltage, and identify a main harmonic source according to each obtained mutual information value.
The embodiment of the invention has the following beneficial effects:
1. according to the invention, only the harmonic voltage signal of the PCC node is required to be obtained, the harmonic current is estimated by using the FastICA algorithm, and the mutual information is estimated through the mutual information deep learning model to identify the main harmonic source, so that the operation is simple, the position of the harmonic current source can be identified under the condition that the system network parameters are unknown, and the problem that the multi-harmonic source in the complex power distribution network is difficult to effectively identify and position by the existing method is solved;
2. the FastICA algorithm in the invention estimates the harmonic current, so that the convergence speed is high, the separation effect is good, the iteration is stable, the separation of non-Gaussian independent components can be carried out, and the accuracy and the reliability of the harmonic current estimation can be improved;
3. the mutual information deep learning model in the invention has simple calculation, can be better adapted to complex nonlinear relations in the power grid, and has stronger universality and reliability.
It should be noted that, in the above system embodiment, each included system unit is only divided according to the functional logic, but not limited to the above division, so long as the corresponding function can be implemented; in addition, the specific names of the functional units are also only for distinguishing from each other, and are not used to limit the protection scope of the present invention.
Those of ordinary skill in the art will appreciate that all or a portion of the steps in implementing the methods of the above embodiments may be implemented by a program to instruct related hardware, where the program may be stored in a computer readable storage medium, such as ROM/RAM, a magnetic disk, an optical disk, etc.
The foregoing disclosure is illustrative of the present invention and is not to be construed as limiting the scope of the invention, which is defined by the appended claims.
Claims (9)
1. A method for identifying multiple harmonic sources of a power distribution network, the method comprising the steps of:
acquiring harmonic voltage signals generated by superposition of a plurality of harmonic sources on PCC nodes, and measuring harmonic voltage values;
separating a fast variation component and a slow variation component in the harmonic voltage signal, and estimating a harmonic current value injected by the PCC node based on the fast variation component;
and importing the obtained harmonic current value and the harmonic voltage value into a pre-trained mutual information deep learning model to obtain a mutual information value between the harmonic current emitted by each harmonic source and the harmonic voltage of the PCC node, and identifying a main harmonic source according to each obtained mutual information value.
2. The method of power distribution network multi-harmonic source identification of claim 1 wherein the harmonic voltage signal is passed through a filter to separate the fast-varying component and the slow-varying component.
3. The power distribution network multi-harmonic source identification method as in claim 1 wherein the values of harmonic currents injected by the PCC nodes are implemented by a fastca algorithm; wherein,
the FastICA algorithm includes the steps of: and preprocessing data, and establishing an objective function for optimizing.
4. A power distribution network multi-harmonic source identification method as in claim 3 wherein the data pre-processing comprises a de-centering process and a whitening process; wherein,
the de-centering process refers to subtracting the average value of all the sampled signals to obtain a group of original data with zero average value: wherein ,X(ti ) Is a sampled signal;
the whitening process is a process of decorrelating the sampled data such that the observed signal X has a unit variance: x=ed -1/ 2 E TX； wherein ,consists of n eigenvalues; d, d i The corresponding feature vector is c i ，E＝[c 1 ,c 2 ,...c n ]The method comprises the steps of carrying out a first treatment on the surface of the The observation signal x=as, a is a mixing matrix, and S is a harmonic current source injected into the PCC node.
5. A method of identifying multiple harmonic sources in a power distribution network as claimed in claim 3 wherein the process of creating an objective function for optimization is to find a direction such that the output w T X(y＝w T X) maximum non-gaussian; wherein,
non-gaussian properties are measured by an approximation of negative entropy: n (N) g (Y)＝{E[g(Y)]-E[g(Y Gauss )]} 2 I.e. find J G (w)＝[E{G(w T X)}] 2 A maximum value; where W is an m-dimensional variable representing one row of the unmixed matrix W;
the objective function is defined as:according to the Kunhn-Tucker condition, the problem of unconstrained optimization is translated, so that the objective function is transformed into: f (w)＝E[G(w T X)]+C(||w|| 2 -1); the optimal solution of the objective function is obtained through a Newton iteration method: />
6. The power distribution network multi-harmonic source identification method as claimed in claim 5 wherein the mutual information deep learning model is constructed based on a neural network; wherein,
the neural network comprises an input layer, a plurality of hidden layers and an output layer; the input layer receives samples of the harmonic current value Y and the harmonic voltage value X as input, and the hidden layer obtains a result Z of the output layer after a series of nonlinear transformation.
7. The method for identifying multiple harmonic sources of a power distribution network according to claim 6, wherein the neural network is trained by performing the following steps, specifically comprising:
7.1 initializing neural network parameters;
7.2 extracting b miniband samples from the joint distribution; wherein the b miniband samples extracted in the joint distribution are denoted (x) (1) ,y (1) ),...,(x (b) ,y (b) ) Obeying joint probability distribution
7.3 extracting b samples from the Y edge distribution; wherein the b samples extracted from the Y edge distribution are recorded asObeying the joint probability distribution->
7.4, evaluating the mutual information lower bound; the mutual information lower bound calculation formula is as follows:
7.5 correcting the deviation correction gradient by EMA; wherein the EMA correction bias correction gradient is expressed as:
7.6, updating parameters of the neural network; wherein the parameters of the updated neural network are
7.7 repeating the steps 7.1 to 7.6 until the convergence condition is reached.
8. The power distribution network multi-harmonic source identification method as in claim 7 wherein the lower bound of mutual information between X and Y is calculated using KL divergence calculated using dual form:
where T is a function defined on X Y that all makes two expectations finite;and I (X; Y) is not less than I θ (X,Y)；/>Is the amount of neural information and is expressed by referring to the gradient formulaDescent to maximize;
wherein the mutual information deep learning model can be realized byFormula (VI)To express, wherein->The empirical distribution of distribution P given n independent samplings is represented.
9. A power distribution network multi-harmonic source identification system, comprising:
the harmonic voltage acquisition unit is used for acquiring harmonic voltage signals generated by superposition of a plurality of harmonic sources on the PCC node and measuring harmonic voltage values;
the harmonic current estimation unit is used for separating a fast variation component and a slow variation component in the harmonic voltage signal and estimating a harmonic current value injected by the PCC node based on the fast variation component;
the main harmonic source identification unit is used for leading the obtained harmonic current value and the harmonic voltage value into a pre-trained mutual information deep learning model to obtain a mutual information value between the harmonic current emitted by each harmonic source and the harmonic voltage of the PCC node, and identifying a main harmonic source according to each obtained mutual information value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310573380.7A CN116609612A (en) | 2023-05-19 | 2023-05-19 | Multi-harmonic source identification method and system for power distribution network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310573380.7A CN116609612A (en) | 2023-05-19 | 2023-05-19 | Multi-harmonic source identification method and system for power distribution network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116609612A true CN116609612A (en) | 2023-08-18 |
Family
ID=87677598
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310573380.7A Pending CN116609612A (en) | 2023-05-19 | 2023-05-19 | Multi-harmonic source identification method and system for power distribution network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116609612A (en) |
-
2023
- 2023-05-19 CN CN202310573380.7A patent/CN116609612A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110728360B (en) | Micro-energy device energy identification method based on BP neural network | |
| CN110210644A (en) | The traffic flow forecasting method integrated based on deep neural network | |
| CN108957421B (en) | Radar radiation source individual identification method and system based on Bezier curve fitting | |
| CN109085469A (en) | A kind of method and system of the signal type of the signal of cable local discharge for identification | |
| CN110147760B (en) | Novel efficient electric energy quality disturbance image feature extraction and identification method | |
| CN110726898B (en) | Power distribution network fault type identification method | |
| CN110796120A (en) | Time domain feature-based circuit breaker mechanical fault XGboost diagnosis method | |
| CN114114039A (en) | Method and device for evaluating consistency of single battery cells of battery system | |
| CN112967210B (en) | Unmanned aerial vehicle image denoising method based on full convolution twin network | |
| CN112347910B (en) | Signal fingerprint identification method based on multi-mode deep learning | |
| CN116911169A (en) | EDFA test parameter prediction method based on BP neural network | |
| CN116527346A (en) | Threat node perception method based on deep learning graph neural network theory | |
| CN114841266A (en) | Voltage sag identification method based on triple prototype network under small sample | |
| CN113095353B (en) | Underdetermined blind source separation method based on AP clustering | |
| CN114218988A (en) | Method for identifying unidirectional ground fault feeder line under unbalanced samples | |
| CN116609612A (en) | Multi-harmonic source identification method and system for power distribution network | |
| CN117236022A (en) | Training method and application method of residual life prediction model of transformer and electronic equipment | |
| CN112738724B (en) | Method, device, equipment and medium for accurately identifying regional target crowd | |
| CN113011597B (en) | Deep learning method and device for regression task | |
| CN113420733B (en) | Efficient distributed big data acquisition implementation method and system | |
| Zhang et al. | A novel analog circuit fault diagnosis approach | |
| CN115293090A (en) | Multi-harmonic source responsibility quantification method based on reconstruction data processing | |
| CN112285541B (en) | Fault diagnosis method for current frequency conversion circuit | |
| CN114755529A (en) | Multi-feature fusion single-phase earth fault type identification method based on deep learning | |
| CN110826573B (en) | Saliency map fusion method and system |
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 |