CN116578855A - Electric energy metering method, system, equipment and medium of alternating-current micro-grid - Google Patents
Electric energy metering method, system, equipment and medium of alternating-current micro-grid Download PDFInfo
- Publication number
- CN116578855A CN116578855A CN202310553477.1A CN202310553477A CN116578855A CN 116578855 A CN116578855 A CN 116578855A CN 202310553477 A CN202310553477 A CN 202310553477A CN 116578855 A CN116578855 A CN 116578855A
- Authority
- CN
- China
- Prior art keywords
- white noise
- target
- noise data
- electric energy
- data set
- 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 39
- 238000007781 pre-processing Methods 0.000 claims abstract description 20
- 238000012549 training Methods 0.000 claims description 90
- 238000000354 decomposition reaction Methods 0.000 claims description 27
- 230000006870 function Effects 0.000 claims description 16
- 230000009466 transformation Effects 0.000 claims description 13
- 238000004590 computer program Methods 0.000 claims description 11
- 238000011156 evaluation Methods 0.000 claims description 9
- 230000004044 response Effects 0.000 claims description 9
- 238000010606 normalization Methods 0.000 claims description 7
- 238000003860 storage Methods 0.000 claims description 6
- 238000010276 construction Methods 0.000 claims description 4
- 238000005259 measurement Methods 0.000 claims description 3
- 238000010248 power generation Methods 0.000 abstract description 20
- 230000010355 oscillation Effects 0.000 abstract description 8
- 238000001514 detection method Methods 0.000 description 10
- 238000005070 sampling Methods 0.000 description 10
- 238000001228 spectrum Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 7
- 238000004146 energy storage Methods 0.000 description 7
- 238000012545 processing Methods 0.000 description 7
- 239000000243 solution Substances 0.000 description 7
- 238000012360 testing method Methods 0.000 description 7
- 230000009286 beneficial effect Effects 0.000 description 5
- 239000002131 composite material Substances 0.000 description 3
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000002474 experimental method Methods 0.000 description 3
- 238000004891 communication Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000005611 electricity Effects 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000002347 injection Methods 0.000 description 2
- 239000007924 injection Substances 0.000 description 2
- 230000009191 jumping Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000000926 separation method Methods 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 238000010835 comparative analysis Methods 0.000 description 1
- 238000009795 derivation Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000029087 digestion Effects 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000005284 excitation Effects 0.000 description 1
- 238000011065 in-situ storage Methods 0.000 description 1
- 238000002955 isolation Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- QHGUCRYDKWKLMG-UHFFFAOYSA-N octopamine Chemical compound NCC(O)C1=CC=C(O)C=C1 QHGUCRYDKWKLMG-UHFFFAOYSA-N 0.000 description 1
- 229960001576 octopamine Drugs 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000011056 performance test Methods 0.000 description 1
- 239000000047 product Substances 0.000 description 1
- 230000000452 restraining effect Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
- 238000009827 uniform distribution Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R22/00—Arrangements for measuring time integral of electric power or current, e.g. electricity meters
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/02—Preprocessing
- G06F2218/04—Denoising
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/08—Feature extraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/12—Classification; Matching
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Power Engineering (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention discloses an electric energy metering method, system, equipment and medium of an alternating-current micro-grid, which are used for responding to a received electric energy metering request, determining an alternating-current micro-grid to be detected corresponding to the electric energy metering request, acquiring a corresponding power grid harmonic data set, preprocessing the power grid harmonic data set, constructing a corresponding target white noise data set, inputting a preset initial electric energy metering model to train by adopting the target white noise data set, generating a corresponding target electric energy metering model, inputting the harmonic data to be detected into the target electric energy metering model, and outputting target harmonic electric energy corresponding to the alternating-current micro-grid to be detected; the electric energy metering device solves the technical problems that the existing electric energy metering device operates under standard sinusoidal voltage and current signals with the rated frequency of 50Hz, interference factors such as harmonic waves and inter-harmonic waves in an alternating current micro-grid-connected power generation environment are not considered, signal distortion, subsynchronous oscillation, voltage fluctuation and the like exist, and therefore accuracy of electric energy metering is lowered.
Description
Technical Field
The invention relates to the technical field of electric energy metering, in particular to an electric energy metering method, an electric energy metering system, electric energy metering equipment and an electric energy metering medium for an alternating-current micro-grid.
Background
With the construction of a novel power system taking new energy as a main body, the distributed power supply has become an important supplement for the development of the novel power system due to the advantages of flexible power generation, in-situ digestion, green cleaning and the like. The AC micro-grid is used as an effective way for improving the utilization of the distributed power supply, and has the characteristics of uninterrupted power supply, flexible operation mode, capability of being used for important load emergency power supplies and the like. However, since a large number of nonlinear loads are flexibly connected into the AC micro-grid, the characteristics of randomness, uncertainty and the like are provided, and the electric energy between the user and the AC micro-grid flows bidirectionally, the anti-interference capability of the AC micro-grid system is poor.
At present, the existing electric energy metering device operates under standard sinusoidal voltage and current signals with rated frequency of 50Hz, and interference factors such as harmonic waves and inter-harmonic waves in an alternating current micro-grid-connected power generation environment are not considered, so that the problems of signal distortion, subsynchronous oscillation, voltage fluctuation and the like can be solved, and the accuracy of electric energy metering is lowered.
Disclosure of Invention
The invention provides an electric energy metering method, system, equipment and medium of an alternating-current micro-grid, which solve the technical problems that the existing electric energy metering device operates under standard sinusoidal voltage and current signals with rated frequency of 50Hz, and interference factors such as harmonic waves and inter-harmonic waves in a grid-connected power generation environment of the alternating-current micro-grid are not considered, and the problems of signal distortion, subsynchronous oscillation, voltage fluctuation and the like exist, so that the accuracy of electric energy metering is reduced.
The electric energy metering method of the alternating-current micro-grid provided by the first aspect of the invention comprises the following steps:
responding to a received electric energy metering request, determining an alternating current micro-grid to be detected corresponding to the electric energy metering request, and acquiring a corresponding grid harmonic data set;
preprocessing the power grid harmonic data set to construct a corresponding target white noise data set;
inputting the target white noise data set into a preset initial electric energy metering model for training, and generating a corresponding target electric energy metering model;
and inputting the harmonic data to be detected into the target electric energy metering model, and outputting the target harmonic electric energy corresponding to the AC micro-grid to be detected.
Optionally, the step of preprocessing the grid harmonic data set to generate a corresponding white noise data set includes:
Respectively injecting a plurality of different types of white noise signals into each power grid harmonic data in the power grid harmonic data set to generate a plurality of initial white noise data corresponding to each power grid harmonic data;
constructing an initial white noise data set by adopting a plurality of initial white noise data;
performing empirical mode decomposition on the initial white noise data set to construct an intermediate white noise data set;
and carrying out Hilbert transformation on the intermediate white noise data set to construct a target white noise data set.
Optionally, the step of performing empirical mode decomposition on the initial white noise data set to construct an intermediate white noise data set includes:
decomposing each piece of initial white noise data in the initial white noise data set to obtain a corresponding target component and a target remainder;
respectively carrying out average value operation on the target components and the target residual items corresponding to the initial white noise data of a plurality of types to obtain a plurality of corresponding target average value data, wherein the target average value data comprises component average values and residual item average values;
zero-mean normalization is carried out on each target mean value data, and corresponding intermediate white noise data are generated;
An intermediate white noise data set is constructed using the plurality of intermediate white noise data.
Optionally, the step of performing hilbert transformation on the intermediate white noise data set to construct a target white noise data set includes:
performing Hilbert transform on each piece of intermediate white noise data in the intermediate white noise data set, and determining a corresponding target instantaneous frequency;
inputting a preset evaluation function by adopting each target instantaneous frequency, and outputting a target score corresponding to each intermediate white noise data;
comparing each target score with a preset standard score;
if the target score is smaller than or equal to the preset standard score, eliminating the intermediate white noise data associated with the target score;
if the target score is larger than the preset standard score, taking the intermediate white noise data associated with the target score as target white noise data;
and constructing a target white noise data set by adopting a plurality of target white noise data.
Optionally, the step of inputting the target white noise data set into a preset initial electric energy metering model for training to generate a corresponding target electric energy metering model includes:
inputting the target white noise data set into a preset initial electric energy metering model for training, and generating corresponding training indexes according to training results;
Calculating a training loss value between the training index and the associated standard index;
comparing the training loss value with a preset standard loss value;
if the training loss value is smaller than or equal to the preset standard loss value, stopping training, and generating a target electric energy metering model;
if the training loss value is larger than the preset standard loss value, adjusting network parameters of the preset initial electric energy metering model according to a preset gradient;
and skipping to execute the step of inputting the target white noise data set into a preset initial electric energy metering model to train and generating a corresponding training index according to a training result until the training loss value is smaller than or equal to the preset standard loss value, so as to generate the target electric energy metering model.
An electric energy metering system of an ac micro-grid according to a second aspect of the present invention includes:
the response module is used for responding to the received electric energy metering request, determining an alternating current micro-grid to be detected corresponding to the electric energy metering request and acquiring a corresponding grid harmonic data set;
the data preprocessing module is used for preprocessing the power grid harmonic data set and constructing a corresponding target white noise data set;
The model training module is used for inputting the target white noise data set into a preset initial electric energy metering model for training to generate a corresponding target electric energy metering model;
and the electric energy metering module is used for inputting the harmonic data to be detected into the target electric energy metering model and outputting the target harmonic electric energy corresponding to the AC micro-grid to be detected.
Optionally, the data preprocessing module includes:
the initial white noise data sub-module is used for respectively injecting a plurality of different types of white noise signals into each power grid harmonic data in the power grid harmonic data set to generate a plurality of initial white noise data corresponding to each power grid harmonic data;
an initial white noise data set sub-module for constructing an initial white noise data set using a plurality of initial white noise data;
the middle white noise data set submodule is used for carrying out empirical mode decomposition on the initial white noise data set to construct a middle white noise data set;
and the target white noise data sub-module is used for carrying out Hilbert transformation on the intermediate white noise data set to construct a target white noise data set.
Optionally, the intermediate white noise data set sub-module includes:
The decomposing unit is used for decomposing each piece of initial white noise data in the initial white noise data set to obtain a corresponding target component and a target remainder;
the target mean value data unit is used for respectively carrying out mean value operation on the target components and the target residual items corresponding to the initial white noise data of a plurality of the same types to obtain a plurality of corresponding target mean value data, wherein the target mean value data comprises a component mean value and a residual item mean value;
the intermediate white noise data unit is used for carrying out zero-mean normalization on each target mean value data to generate corresponding intermediate white noise data;
an intermediate white noise data set construction unit for constructing an intermediate white noise data set using the plurality of intermediate white noise data.
An electronic device according to a third aspect of the present invention includes a memory and a processor, where the memory stores a computer program, and the computer program when executed by the processor causes the processor to execute the steps of the method for measuring electric energy of the ac microgrid according to any one of the above.
A fourth aspect of the present invention provides a computer-readable storage medium having stored thereon a computer program which, when executed, implements a method of metering electrical energy of an ac microgrid as set forth in any one of the preceding claims.
From the above technical scheme, the invention has the following advantages:
responding to a received electric energy metering request, determining an alternating-current micro-grid to be detected corresponding to the electric energy metering request, acquiring a corresponding power grid harmonic data set, preprocessing the power grid harmonic data set, constructing a corresponding target white noise data set, inputting a preset initial electric energy metering model to train by adopting the target white noise data set, generating a corresponding target electric energy metering model, inputting the target electric energy metering model by adopting the harmonic data to be detected, and outputting target harmonic electric energy corresponding to the alternating-current micro-grid to be detected; the electric energy metering device solves the technical problems that the existing electric energy metering device operates under standard sinusoidal voltage and current signals with the rated frequency of 50Hz, and does not consider interference factors such as harmonic waves and inter-harmonic waves in an alternating-current micro-grid-connected power generation environment, and the like, and the problems of signal distortion, subsynchronous oscillation, voltage fluctuation and the like exist, so that the accuracy of electric energy metering is reduced; the harmonic current signals in the alternating-current micro-grid containing nonlinear loads are analyzed through a target electric energy metering model constructed by combining an EEMD decomposition algorithm and a deep belief network (DBN network), namely, local time and frequency characteristic information of the signals is obtained in a self-adaptive mode through the EEMD decomposition algorithm according to the structural characteristics of the harmonic current signals, and the self-adaptive detection of harmonic components is realized by means of strong data representation capability of the DBN network, so that the beneficial effects of accurate detection and metering of harmonic electric energy and inter-harmonic electric energy are achieved.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions of the prior art, the drawings which are used 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 other drawings can be obtained from these drawings without inventive faculty for a person skilled in the art.
Fig. 1 is a flowchart of steps of a method for measuring electric energy of an ac micro-grid according to an embodiment of the present invention;
fig. 2 is a flow chart of steps of a method for measuring electric energy of an ac micro-grid according to a second embodiment of the present invention;
FIG. 3 is a flow chart of harmonic power flow provided by an embodiment of the present invention;
fig. 4 is a schematic diagram of an ac micro-grid according to a second embodiment of the present invention;
FIG. 5 is a schematic diagram of a preset initial power metering model according to a second embodiment of the present invention;
FIG. 6 is a schematic diagram showing the performance of a different target electric energy metering model layer according to a second embodiment of the present invention;
FIG. 7 is a schematic diagram showing an approximation of a model loss function according to a second embodiment of the present invention;
fig. 8 is a diagram of a harmonic adaptive detection architecture according to a second embodiment of the present invention;
Fig. 9 is a block diagram of an electric energy metering system of an ac micro-grid according to a third embodiment of the present invention.
Detailed Description
The embodiment of the invention provides an electric energy metering method, an electric energy metering system, electric energy metering equipment and an electric energy metering medium for an alternating-current micro-grid, which are used for solving the technical problems that the existing electric energy metering device operates under standard sinusoidal voltage and current signals with the rated frequency of 50Hz, interference factors such as harmonic waves and inter-harmonic waves in a grid-connected power generation environment of the alternating-current micro-grid are not considered, and the problems of signal distortion, subsynchronous oscillation, voltage fluctuation and the like exist, so that the accuracy of electric energy metering is reduced.
The metering modes of the electric energy include three modes of full-wave electric energy metering, fundamental wave electric energy metering and fundamental wave and harmonic wave electric energy metering respectively. The full-wave electric energy metering mode can cause that a linear load user in an alternating-current micro-grid pays more electricity fees when suffering from harmonic hazard; and the nonlinear load users not only cause harmonic pollution, but also pay less electricity. The fundamental wave electric energy metering mode can avoid that a linear load user bears extra harmonic wave electric energy cost, but a nonlinear load user only needs to pay fundamental wave active electric energy cost, and the action of generating harmonic wave pollution is not subjected to any economic penalty. The fundamental wave electric energy and harmonic wave electric energy separated metering mode distributes corresponding electric energy cost according to the actual injection or consumption of electric power harmonic wave, thereby being beneficial to more accurately metering electric energy, and having excitation effects on both supply and demand sides, restraining electric power harmonic wave, improving electric energy utilization rate and the like.
The nonlinear load in the AC micro-grid system is used as a harmonic source, and when active power is absorbed, part of fundamental wave power is converted into harmonic waves and inter-harmonic waves and is transmitted to the AC micro-grid system, so that voltage and current waveform distortion of the linear load in the system is caused. The harmonic wave and inter-harmonic wave active power generated by the nonlinear load are opposite to the fundamental wave active power in direction, so that the total active power consumed by the nonlinear load is smaller than the fundamental wave active power, and the electric energy is less. As analyzed above, the linear load consumes more total active power than the fundamental active power, resulting in more electrical energy. The harmonic power flow direction is shown in fig. 3.
The fundamental wave active power balance expression is shown as the formula (1):
P G ＝P 1 +P 2 +…+P n +P s (1)
the harmonic and inter-harmonic active power balance expression is shown in the formula (2):
P K ＝P SK +P 1K +P 2K …+P nK (2)
in order to make the objects, features and advantages of the present invention more comprehensible, the technical solutions in the embodiments of the present invention are described in detail below with reference to the accompanying drawings, and it is apparent that the embodiments described below are only some embodiments of the present invention, but not all embodiments of the present invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1, fig. 1 is a flowchart illustrating a method for measuring electric energy of an ac micro grid according to an embodiment of the invention.
The invention provides an electric energy metering method of an alternating-current micro-grid, which comprises the following steps:
and step 101, responding to the received electric energy metering request, determining an alternating current micro-grid to be detected corresponding to the electric energy metering request, and acquiring a corresponding grid harmonic data set.
The electric energy metering request refers to request information for metering electric energy for an alternating-current micro-grid.
The power grid harmonic data set refers to collecting voltage and current signal data generated by an alternating current micro-grid to be detected under a nonlinear load, and the selected characteristic parameters comprise a data source type, sampling nodes, sampling frequency and data length.
In the embodiment of the invention, an alternating-current micro-grid to be detected corresponding to an electric energy metering request is determined and a corresponding grid harmonic data set is acquired in response to the received electric energy metering request.
Step 102, preprocessing the power grid harmonic data set to construct a corresponding target white noise data set.
Preprocessing, namely injecting a white noise signal into a power grid harmonic data set, then carrying out EMD decomposition and Hilbert transformation, and screening out a target white noise data set with higher score through an evaluation function.
In the embodiment of the invention, a white noise signal is injected into a power grid harmonic data set, EMD decomposition and Hilbert transformation are carried out, and then a target white noise data set with higher score is screened out through an evaluation function.
And step 103, inputting a preset initial electric energy metering model by using the target white noise data set to train, and generating a corresponding target electric energy metering model.
The target electric energy metering model is used for receiving harmonic data corresponding to the AC micro-grid to be detected and outputting target harmonic electric energy corresponding to the AC micro-grid to be detected.
According to the embodiment of the invention, the target white noise data set obtained through preprocessing is input into a preset initial electric energy metering model to train according to the target white noise data set, so that a corresponding target electric energy metering model is generated.
And 104, inputting the harmonic data to be detected into a target electric energy metering model, and outputting the target harmonic electric energy corresponding to the AC micro-grid to be detected.
In the embodiment of the invention, the harmonic data to be detected is input into a target electric energy metering model, corresponding voltage signals, current signals and amplitude spectrums are output, the frequency spectrum peak fitting is performed by adopting a least square method, the amplitude, angular frequency and initial phase angle of the electric power harmonic are obtained, and then the corresponding target harmonic electric energy is generated.
In the method, an alternating-current micro-grid to be detected corresponding to an electric energy metering request is determined in response to the received electric energy metering request, a corresponding power grid harmonic data set is obtained, preprocessing is carried out on the power grid harmonic data set, a corresponding target white noise data set is constructed, a preset initial electric energy metering model is input into the target white noise data set for training, a corresponding target electric energy metering model is generated, the harmonic data to be detected is input into the target electric energy metering model, and target harmonic electric energy corresponding to the alternating-current micro-grid to be detected is output; the electric energy metering device solves the technical problems that the existing electric energy metering device operates under standard sinusoidal voltage and current signals with the rated frequency of 50Hz, and does not consider interference factors such as harmonic waves and inter-harmonic waves in an alternating-current micro-grid-connected power generation environment, and the like, and the problems of signal distortion, subsynchronous oscillation, voltage fluctuation and the like exist, so that the accuracy of electric energy metering is reduced; the harmonic current signals in the alternating-current micro-grid containing nonlinear loads are analyzed through a target electric energy metering model constructed by combining an EEMD decomposition algorithm and a deep belief network (DBN network), namely, local time and frequency characteristic information of the signals is obtained in a self-adaptive mode through the EEMD decomposition algorithm according to the structural characteristics of the harmonic current signals, and the self-adaptive detection of harmonic components is realized by means of strong data representation capability of the DBN network, so that the beneficial effects of accurate detection and metering of harmonic electric energy and inter-harmonic electric energy are achieved.
Referring to fig. 2, fig. 2 is a flowchart illustrating a method for measuring electric energy of an ac micro-grid according to a second embodiment of the present invention.
The invention provides an electric energy metering method of an alternating-current micro-grid, which comprises the following steps:
step 201, in response to a received electric energy metering request, determining an alternating current micro-grid to be detected corresponding to the electric energy metering request and acquiring a corresponding grid harmonic data set.
In this embodiment, the ac micro-grid is formed by photovoltaic power generation with a rated power of 150kW, wind power generation with a rated power of 15kW, a hybrid energy storage system with a rated power of 80kW and an energy management system, and is connected to a large power grid through a 400V ac bus, as shown in fig. 4.
The electric energy meter data acquisition equipment is a DEWE5000 type electric energy quality tester, and is provided with 16 isolation input modules and a 100Mb/s signal output rate, so that voltage and current signal field data acquisition can be realized. The acquired original data are voltage and current signals, the number of the obtained total sample points is 3×106, the fundamental wave frequency is 50Hz, 200 points are sampled in one cycle, and 60000 cycles are sampled in 5 min. The sampling point is arranged at a PCC point of an alternating current MG laboratory, different working conditions are switched through switch control, and the acquisition range comprises real-time voltage and current data under working conditions such as energy storage discharge, wind power generation, photovoltaic power generation, wind-solar energy storage and the like.
The power grid harmonic data set comprises voltage and current signals generated by nonlinear loads such as energy storage discharge, wind power generation, photovoltaic power generation, wind-light energy storage power generation and the like, and the selected characteristic parameters comprise data source types, sampling nodes, sampling frequency and data length. Firstly segmenting data, taking 100000 pieces of data from each group of measurement data, and dividing the data into 2 groups, wherein 1 group is used as a training data set, and the total of 90000 pieces of data are used for training a DBN-EEMD model; the other 1 set was used as a test dataset for validating the model for a total of 10000 strips.
TABLE 1 data Source under different conditions
| Working conditions of | Type(s) | Sampling node | Sampling frequency (kHz) | Length of time |
| Working condition 1 | Energy storage discharge | Grid-connected point | 10kHz | 5min |
| Working condition 2 | Wind power generation | Grid-connected point | 10kHz | 5min |
| Working condition 3 | Photovoltaic power generation | Grid-connected point | 10kHz | 5min |
| Working condition 4 | Wind-solar power generation | Grid-connected point | 10kHz | 5min |
| Working condition 5 | Wind-solar energy storage power generation | Grid-connected point | 10kHz | 5min |
In the embodiment of the present invention, the implementation process of step 201 is similar to that of step 101, and will not be repeated here.
Step 202, injecting a plurality of different types of white noise signals into each power grid harmonic data in the power grid harmonic data set respectively, and generating a plurality of initial white noise data corresponding to each power grid harmonic data.
It is worth mentioning that, aiming at the problems of modal aliasing, frequency fluctuation and the like which are easy to occur when the EMD signal is decomposed, the EEMD method is proposed by Norden E.Huang in combination with the uniform distribution and zero mean characteristic of Gaussian white noise (White Gaussian Noise, WGN) frequency spectrum. Due to the injection of the WGN signal, the micro-grid output voltage and current signals of different time scales are automatically distributed on the appropriate reference scales. The injected WGN signals cancel each other out through multiple decompositions. The average of the EEMD decomposition is taken as the final result.
It is worth mentioning that the original voltage-current signal is decomposed by EEMD, and only the first four IMF components are considered, since the information of the electric energy metering is mainly represented at high frequency.
In the embodiment of the invention, the power grid harmonic data are voltage and current signals, and the power grid harmonic data x (t) generated by the nonlinear load under different working conditions are injected into white noise omega i In (t), a composite signal X is obtained i (t) the formula is as follows:
X i (t)＝x(t)+ω i (t) (3)
where x (t) represents the ith grid harmonic data, ω i (t) represents the ith white noise, X i (t) represents the ith composite signal, i=1, 2, …, m.
Step 203, constructing an initial white noise data set by using a plurality of initial white noise data.
In the embodiment of the invention, a plurality of initial white noise data sets are adopted to construct an initial white noise data set.
And 204, performing empirical mode decomposition on the initial white noise data set to construct an intermediate white noise data set.
Further, step 204 may include the sub-steps of:
s11, decomposing each initial white noise data in the initial white noise data set to obtain a corresponding target component and a target remainder.
In the embodiment of the invention, the composite signal X i (t) performing EMD decomposition to obtain n IMF (Intrinsic Mode Function, IMF) target components c i (t) and target remainder r i (t) formulae such asThe following steps:
X i (t)＝c i (t)+r i (t) (4)
wherein, c i (t) represents a target component, r i (t) represents a target remainder, i=1, 2, …, n.
And S12, respectively carrying out average value operation on target components and target remainder corresponding to a plurality of initial white noise data of the same type to obtain a plurality of corresponding target average value data, wherein the target average value data comprises a component average value and a remainder average value.
It should be noted that in step 202, different WGN signals are injected into each grid harmonic data x (t).
In the embodiment of the invention, by utilizing the zero-mean principle of Gaussian white noise spectrum, different WGN signals are injected into the power grid harmonic data x (t) in the step 202, the formula (3) and the formula (4) are repeated for reaching preset decomposition times, and the decomposed IMF components are averaged, wherein the formula is as follows:
Where c (t) represents the component mean, n represents the IMF component number, r (t) represents the remainder mean, and m represents the remainder component number.
It should be noted that each piece of initial white noise data corresponds to a component mean value and a residual mean value, where there are a plurality of pieces of initial white noise data.
S13, carrying out zero-mean normalization on each target mean value data to generate corresponding intermediate white noise data.
In the embodiment of the invention, the WGN signal zero mean normalization is utilized to eliminate the WGN effect and obtain the intermediate white noise data x s (t) the formula is as follows:
x s (t) ＝c(t)+r(t) (7)
wherein x is s (t) representsIntermediate white noise data.
S14, constructing an intermediate white noise data set by adopting a plurality of intermediate white noise data.
In an embodiment of the invention, a plurality of intermediate white noise data sets are employed to construct an intermediate white noise data set.
Step 205, performing hilbert transformation on the intermediate white noise data set to construct a target white noise data set.
S21, performing Hilbert transform on each middle white noise data in the middle white noise data set, and determining a corresponding target instantaneous frequency.
In the embodiment of the invention, the intermediate white noise data x s (t) resolved intermediate IMF component c s (t) performing Hilbert transform to obtain a target instantaneous frequency, wherein the formula is as follows:
It is worth mentioning that intermediate white noise data x s (t) resolved intermediate IMF component c s (t) performing Hilbert transformation to perform Hilbert transformation, so as to obtain an analytic signal; the instantaneous angular frequency function and the instantaneous frequency function are obtained by the phase function.
Wherein f i (t) represents the ith target instantaneous frequency, ω (t) represents the instantaneous angular frequency,representing phase, alpha (t) representing Hilbert transform resolved signal, H [ c ] s (t)]Representing amplitude, c s (t) represents the intermediate IMF component, z (t) represents the resolved signal, and j represents the experimental standard deviation.
It should be noted that, after the analysis signal z (t) is transformed, the amplitude spectrum H (ω (t), t) can be obtained, and according to the above derivation, the intermediate white noise data x in the ac MG can be obtained s The Hilbert marginal energy spectrum E (ω (t)) of (t) reflecting the total amplitude of a certain frequency of the current signal and the energy distribution over the time span, as shown in equation (14):
H(ω(t),t)＝Re∑α(t)e j∫ω(t)dt (13)
where H (ω (t), t) represents the magnitude spectrum, E (ω (t)) represents the marginal energy spectrum, and Re represents the Fourier transform.
In addition, when EEMD decomposition is performed, the added WGN should be properly set with coefficients, and if the coefficients are too large or too small, the decomposition error will be increased, resulting in meaningless decomposition results. According to the original current signal characteristics in the alternating-current MG, an optimal WGN signal is selected by adopting an evaluation function experiment standard deviation, and the influence on a decomposition result due to unreasonable WGN coefficient setting is reduced.
S21, inputting a preset evaluation function by adopting each target instantaneous frequency, and outputting target scores corresponding to each intermediate white noise data.
In the embodiment of the invention, the instantaneous frequency experimental standard deviation s of the jth IMF is calculated according to the WGN signals added in different times i Obtaining the formula (15):
wherein s is i Represents the standard deviation of the experiment, f i (t)' represents the target instantaneous frequency average of the ith IMF, i=1, 2, …, j, with the smaller sj being the higher the score, depending on the addition of different WGN signals.
S22, comparing each target score with a preset standard score.
In the embodiment of the invention, the target score associated with each intermediate white noise data is compared with the preset standard score
S23, if the target score is smaller than or equal to the preset standard score, eliminating the intermediate white noise data associated with the target score.
In the embodiment of the invention, if the target score is smaller than or equal to the preset standard score, eliminating the intermediate white noise data associated with the target score.
And S24, if the target score is larger than the preset standard score, taking the intermediate white noise data associated with the target score as target white noise data.
In the embodiment of the invention, if the target score is greater than the preset standard score, the intermediate white noise data associated with the target score is used as the target white noise data.
S25, constructing a target white noise data set by adopting a plurality of target white noise data.
In the embodiment of the invention, a plurality of target white noise data sets are adopted to construct the target white noise data set, so that the target white noise data with higher scores are screened out.
And 206, inputting a preset initial electric energy metering model by using the target white noise data set to train, and generating a corresponding target electric energy metering model.
It is worth mentioning that the preset initial electric energy metering model adopts a deep belief network (Deep Belief Network, DBN), the DBN is formed by stacking a plurality of limited Boltzmann machines (Restricted Boltzmann Machine, RBM), and each RBM is trained from a low layer to a high layer in an unsupervised mode. The Back Propagation (BP) algorithm is used to fine tune the training error of the entire pre-set initial power metering model. The preset initial power metering model (network training model of DBN-BP) is shown in fig. 5.
It is worth mentioning that the invention adopts the contrast divergence (Contrastive Divergence, CD) algorithm proposed by Hinton to quickly train and learn the RBM, uses the target white noise data set with the optimal evaluation function in the formula (15) as a label value, uses the RBM output characteristic vector as the input characteristic vector of the preset initial electric energy metering model, carries out supervision training through the preset initial electric energy metering model, reversely and continuously adjusts parameters, and reduces training errors. And finally, obtaining a matched optimal WGN signal by using the target electric energy metering model, and reducing artificial setting errors so that the separation result is more approximate to a theoretical value.
S31, inputting a target white noise data set into a preset initial electric energy metering model for training, and generating corresponding training indexes according to training results.
S32, calculating a training loss value between the training index and the associated standard index.
S33, comparing the training loss value with a preset standard loss value.
And S34, stopping training if the training loss value is smaller than or equal to the preset standard loss value, and generating a target electric energy metering model.
And S35, if the training loss value is larger than the preset standard loss value, adjusting network parameters of a preset initial electric energy metering model according to a preset gradient.
S36, performing jump execution, namely inputting a target white noise data set into a preset initial electric energy metering model for training, and generating a corresponding training index according to a training result until a training loss value is smaller than or equal to a preset standard loss value, so as to generate the target electric energy metering model.
In the embodiment of the invention, a target white noise data set is input into a preset initial electric energy metering model for training, a corresponding training index is generated according to a training result, a training loss value between the training index and an associated standard index is calculated, the training loss value is compared with the preset standard loss value, and if the training loss value is smaller than or equal to the preset standard loss value, training is stopped, so that a target electric energy metering model is generated; and if the training loss value is larger than the preset standard loss value, performing network parameter jump of the preset initial electric energy metering model according to the preset gradient adjustment, inputting a target white noise data set into the preset initial electric energy metering model for training, and generating a corresponding training index according to a training result until the training loss value is smaller than or equal to the preset standard loss value, so as to generate the target electric energy metering model.
It is worth mentioning that, in the verification of the target electric energy metering model, a group is randomly extracted from the harmonic data set of the power grid at a time to verify the accuracy of the model. After the process is finished, randomly extracting data from the data collected under 5 different working conditions, and analyzing the harmonic characteristics of current and power of the data. The data acquisition equipment is provided with the same sampling frequency of 10kHz and the time length of 5min, and is used for data consistency and comparative analysis, and the data sources under different working conditions are shown in Table 2.
And fine tuning the training error of the whole target electric energy metering model by adopting a BP algorithm. The structure of the target electric energy metering model is a key of signal extraction efficiency, and the extraction efficiency of different structures on signal characteristics is compared by quantifying reconstruction errors of different layers of the target electric energy metering model. After multiple experiments, the number of layers of the target electric energy metering model is set to be 3, the number of nodes of each layer is 350, 200 and 50 in sequence, the obtained reconstruction error is minimum (0.010), the time is 8.1 seconds, and the comparison result is shown in Table 2.
Table 2, target electric energy metering model layer number comparison table
It should be noted that, after the loss function of the target electric energy metering model is tested, after 8 iterations, the test error and the training error are respectively reduced to 0.01% and 0.02%, the test shows that the test training error is obviously smaller than the training error, the test error and the training error are basically approximate to zero values, and the performance test results are shown in fig. 6 and 7.
And 207, inputting the harmonic data to be detected into a target electric energy metering model, and outputting the target harmonic electric energy corresponding to the AC micro-grid to be detected.
In the embodiment of the invention, harmonic data to be detected is adopted to be input into a target electric energy metering model, harmonic voltage and harmonic current corresponding to an alternating current micro-grid to be detected are output, and the power of harmonic is calculated:
P＝U(t)·I(t) (17)
where P represents harmonic power, U (t) represents harmonic voltage, and I (t) represents harmonic current.
It is worth mentioning that the fundamental wave power and the harmonic wave power are opposite in direction, the user load generates harmonic waves, the on-site electric energy meter can be used for counting less electric quantity, and therefore the reverse harmonic wave power is independently calculated and analyzed to obtain electric energy generated by harmonic waves:
E＝PT (18)
where E represents the power generated by the inverted harmonic and T represents time.
As shown in fig. 8, the test equipment is used to sample the original harmonic data generated by the nonlinear load under different working conditions in the ac MG, and the sampled data are reasonably classified and integrated into a power grid harmonic data set without labels. Then, white noise signals are artificially added to each group of original data, EMD decomposition and Hilbert transformation are carried out, and then a target white noise data set with higher score is screened out through an evaluation function. And then training a preset initial electric energy metering model by using the original sampling data set and the target white noise data set as test data sets. And finally, when harmonic detection is carried out, directly sending the original sampling data into a trained target electric energy metering model, wherein the target electric energy metering model can be adaptively matched with a white noise signal according to the signal data characteristics in the alternating current MG, so that the artificial setting error is reduced, and the separation result is more approximate to a theoretical value.
In the method, an alternating-current micro-grid to be detected corresponding to an electric energy metering request is determined in response to the received electric energy metering request, a corresponding power grid harmonic data set is obtained, preprocessing is carried out on the power grid harmonic data set, a corresponding target white noise data set is constructed, a preset initial electric energy metering model is input into the target white noise data set for training, a corresponding target electric energy metering model is generated, the harmonic data to be detected is input into the target electric energy metering model, and target harmonic electric energy corresponding to the alternating-current micro-grid to be detected is output; the electric energy metering device solves the technical problems that the existing electric energy metering device operates under standard sinusoidal voltage and current signals with the rated frequency of 50Hz, and does not consider interference factors such as harmonic waves and inter-harmonic waves in an alternating-current micro-grid-connected power generation environment, and the like, and the problems of signal distortion, subsynchronous oscillation, voltage fluctuation and the like exist, so that the accuracy of electric energy metering is reduced; the harmonic current signals in the alternating-current micro-grid containing nonlinear loads are analyzed through a target electric energy metering model constructed by combining an EEMD decomposition algorithm and a deep belief network (DBN network), namely, local time and frequency characteristic information of the signals is obtained in a self-adaptive mode through the EEMD decomposition algorithm according to the structural characteristics of the harmonic current signals, and the self-adaptive detection of harmonic components is realized by means of strong data representation capability of the DBN network, so that the beneficial effects of accurate detection and metering of harmonic electric energy and inter-harmonic electric energy are achieved.
Referring to fig. 9, fig. 9 is a block diagram illustrating an electric energy metering system of an ac micro-grid according to a third embodiment of the present invention.
The invention provides an electric energy metering system of an alternating-current micro-grid, which comprises the following components:
the response module 301 is configured to determine an ac micro-grid to be detected corresponding to the electric energy metering request and obtain a corresponding grid harmonic data set in response to the received electric energy metering request;
the data preprocessing module 302 is configured to preprocess the power grid harmonic data set and construct a corresponding target white noise data set;
the model training module 303 is configured to input a preset initial electric energy metering model to perform training by using a target white noise data set, and generate a corresponding target electric energy metering model;
the electric energy metering module 304 is configured to input the harmonic data to be detected into the target electric energy metering model, and output the target harmonic electric energy corresponding to the ac micro-grid to be detected.
Further, the data preprocessing module 302 includes:
the initial white noise data sub-module is used for respectively injecting a plurality of different types of white noise signals into each power grid harmonic data in the power grid harmonic data set to generate a plurality of initial white noise data corresponding to each power grid harmonic data.
An initial white noise data set sub-module for constructing an initial white noise data set using a plurality of initial white noise data.
And the middle white noise data set sub-module is used for carrying out empirical mode decomposition on the initial white noise data set and constructing the middle white noise data set.
And the target white noise data sub-module is used for carrying out Hilbert transformation on the intermediate white noise data set and constructing a target white noise data set.
Further, the intermediate white noise data set sub-module includes:
the decomposing unit is used for decomposing each initial white noise data in the initial white noise data set to obtain a corresponding target component and a target remainder;
the target mean value data unit is used for respectively carrying out mean value operation on target components and target residual items corresponding to a plurality of initial white noise data of the same type to obtain a plurality of corresponding target mean value data, wherein the target mean value data comprises a component mean value and a residual item mean value;
the intermediate white noise data unit is used for carrying out zero-mean normalization on each target mean value data to generate corresponding intermediate white noise data;
an intermediate white noise data set construction unit for constructing an intermediate white noise data set using the plurality of intermediate white noise data.
Further, the target white noise data submodule includes:
and the target instantaneous frequency unit is used for performing Hilbert transformation on each piece of intermediate white noise data in the intermediate white noise data set and determining the corresponding target instantaneous frequency.
And the target scoring unit is used for inputting a preset evaluation function by adopting each target instantaneous frequency and outputting target scores corresponding to each intermediate white noise data.
And the score comparison unit is used for comparing each target score with a preset standard score.
And the first data processing unit is used for eliminating the intermediate white noise data associated with the target score if the target score is smaller than or equal to the preset standard score.
And the second data processing unit is used for taking the intermediate white noise data associated with the target score as target white noise data if the target score is larger than a preset standard score.
A unit for constructing a target white noise dataset using a plurality of target white noise data.
Further, the model training module 303 includes:
the training index sub-module is used for inputting a preset initial electric energy metering model to train by adopting a target white noise data set, and generating a corresponding training index according to a training result.
And the training loss value submodule is used for calculating the training loss value between the training index and the associated standard index.
And the loss value comparison sub-module is used for comparing the training loss value with a preset standard loss value.
And the first model processing submodule is used for stopping training and generating a target electric energy metering model if the training loss value is smaller than or equal to the preset standard loss value.
And the second model processing sub-module is used for adjusting network parameters of the preset initial electric energy metering model according to the preset gradient if the training loss value is larger than the preset standard loss value.
And the rotor jumping module is used for jumping and executing the step of inputting a target white noise data set into a preset initial electric energy metering model for training and generating a corresponding training index according to a training result until the training loss value is smaller than or equal to a preset standard loss value to generate the target electric energy metering model.
In the method, an alternating-current micro-grid to be detected corresponding to an electric energy metering request is determined in response to the received electric energy metering request, a corresponding power grid harmonic data set is obtained, preprocessing is carried out on the power grid harmonic data set, a corresponding target white noise data set is constructed, a preset initial electric energy metering model is input into the target white noise data set for training, a corresponding target electric energy metering model is generated, the harmonic data to be detected is input into the target electric energy metering model, and target harmonic electric energy corresponding to the alternating-current micro-grid to be detected is output; the electric energy metering device solves the technical problems that the existing electric energy metering device operates under standard sinusoidal voltage and current signals with the rated frequency of 50Hz, and does not consider interference factors such as harmonic waves and inter-harmonic waves in an alternating-current micro-grid-connected power generation environment, and the like, and the problems of signal distortion, subsynchronous oscillation, voltage fluctuation and the like exist, so that the accuracy of electric energy metering is reduced; the harmonic current signals in the alternating-current micro-grid containing nonlinear loads are analyzed through a target electric energy metering model constructed by combining an EEMD decomposition algorithm and a deep belief network (DBN network), namely, local time and frequency characteristic information of the signals is obtained in a self-adaptive mode through the EEMD decomposition algorithm according to the structural characteristics of the harmonic current signals, and the self-adaptive detection of harmonic components is realized by means of strong data representation capability of the DBN network, so that the beneficial effects of accurate detection and metering of harmonic electric energy and inter-harmonic electric energy are achieved.
An electronic device according to an embodiment of the present invention includes: a memory and a processor, the memory storing a computer program; the computer program, when executed by a processor, causes the processor to perform the method of metering electrical energy of an ac microgrid according to any of the embodiments described above.
The memory may be an electronic memory such as a flash memory, an EEPROM (electrically erasable programmable read only memory), an EPROM, a hard disk, or a ROM. The memory has memory space for program code to perform any of the method steps described above. For example, the memory space for the program code may include individual program code for implementing the various steps in the above method, respectively. The program code can be read from or written to one or more computer program products. These computer program products comprise a program code carrier such as a hard disk, a Compact Disc (CD), a memory card or a floppy disk. The program code may be compressed, for example, in a suitable form. The code, when executed by a computing processing device, causes the computing processing device to perform the steps in the method described above.
An embodiment of the present application provides a computer-readable storage medium having stored thereon a computer program that, when executed, implements a method for metering electrical energy of an ac microgrid according to any embodiment of the present application.
It will be clear to those skilled in the art that, for convenience and brevity of description, specific working procedures of the above-described systems, apparatuses and units may refer to corresponding procedures in the foregoing method embodiments, which are not repeated herein.
In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices, and methods may be implemented in other manners. For example, the apparatus embodiments described above are merely illustrative, e.g., the division of the units is merely a logical function division, and there may be additional divisions when actually implemented, e.g., multiple units or components may be combined or integrated into another system, or some features may be omitted or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection via some interfaces, devices or units, which may be in electrical, mechanical or other form.
The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in the embodiments of the present invention may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units.
The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied essentially or in part or all of the technical solution or in part in the form of a software product stored in a storage medium, including instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
The above embodiments are only for illustrating the technical solution of the present invention, and not for limiting the same; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (10)
1. An electrical energy metering method for an ac micro-grid, comprising:
responding to a received electric energy metering request, determining an alternating current micro-grid to be detected corresponding to the electric energy metering request, and acquiring a corresponding grid harmonic data set;
preprocessing the power grid harmonic data set to construct a corresponding target white noise data set;
inputting the target white noise data set into a preset initial electric energy metering model for training, and generating a corresponding target electric energy metering model;
and inputting the harmonic data to be detected into the target electric energy metering model, and outputting the target harmonic electric energy corresponding to the AC micro-grid to be detected.
2. The method of ac microgrid power metering according to claim 1, wherein said step of preprocessing said grid harmonic data set to generate a corresponding white noise data set comprises:
respectively injecting a plurality of different types of white noise signals into each power grid harmonic data in the power grid harmonic data set to generate a plurality of initial white noise data corresponding to each power grid harmonic data;
constructing an initial white noise data set by adopting a plurality of initial white noise data;
performing empirical mode decomposition on the initial white noise data set to construct an intermediate white noise data set;
and carrying out Hilbert transformation on the intermediate white noise data set to construct a target white noise data set.
3. The method for power metering of an ac micro-grid according to claim 2, wherein the step of performing empirical mode decomposition on the initial white noise data set to construct an intermediate white noise data set comprises:
decomposing each piece of initial white noise data in the initial white noise data set to obtain a corresponding target component and a target remainder;
respectively carrying out average value operation on the target components and the target residual items corresponding to the initial white noise data of a plurality of types to obtain a plurality of corresponding target average value data, wherein the target average value data comprises component average values and residual item average values;
Zero-mean normalization is carried out on each target mean value data, and corresponding intermediate white noise data are generated;
an intermediate white noise data set is constructed using the plurality of intermediate white noise data.
4. The method of power metering for an ac microgrid according to claim 3, wherein said step of performing a hilbert transform on said intermediate white noise data set to construct a target white noise data set comprises:
performing Hilbert transform on each piece of intermediate white noise data in the intermediate white noise data set, and determining a corresponding target instantaneous frequency;
inputting a preset evaluation function by adopting each target instantaneous frequency, and outputting a target score corresponding to each intermediate white noise data;
comparing each target score with a preset standard score;
if the target score is smaller than or equal to the preset standard score, eliminating the intermediate white noise data associated with the target score;
if the target score is larger than the preset standard score, taking the intermediate white noise data associated with the target score as target white noise data;
and constructing a target white noise data set by adopting a plurality of target white noise data.
5. The method for measuring electric energy of an ac micro-grid according to claim 1, wherein the step of inputting the target white noise data set into a preset initial electric energy measurement model for training to generate a corresponding target electric energy measurement model comprises the steps of:
Inputting the target white noise data set into a preset initial electric energy metering model for training, and generating corresponding training indexes according to training results;
calculating a training loss value between the training index and the associated standard index;
comparing the training loss value with a preset standard loss value;
if the training loss value is smaller than or equal to the preset standard loss value, stopping training, and generating a target electric energy metering model;
if the training loss value is larger than the preset standard loss value, adjusting network parameters of the preset initial electric energy metering model according to a preset gradient;
and skipping to execute the step of inputting the target white noise data set into a preset initial electric energy metering model to train and generating a corresponding training index according to a training result until the training loss value is smaller than or equal to the preset standard loss value, so as to generate the target electric energy metering model.
6. An electrical energy metering system for an ac microgrid, comprising:
the response module is used for responding to the received electric energy metering request, determining an alternating current micro-grid to be detected corresponding to the electric energy metering request and acquiring a corresponding grid harmonic data set;
The data preprocessing module is used for preprocessing the power grid harmonic data set and constructing a corresponding target white noise data set;
the model training module is used for inputting the target white noise data set into a preset initial electric energy metering model for training to generate a corresponding target electric energy metering model;
and the electric energy metering module is used for inputting the harmonic data to be detected into the target electric energy metering model and outputting the target harmonic electric energy corresponding to the AC micro-grid to be detected.
7. The ac microgrid power metering system according to claim 6, wherein said data preprocessing module comprises:
the initial white noise data sub-module is used for respectively injecting a plurality of different types of white noise signals into each power grid harmonic data in the power grid harmonic data set to generate a plurality of initial white noise data corresponding to each power grid harmonic data;
an initial white noise data set sub-module for constructing an initial white noise data set using a plurality of initial white noise data;
the middle white noise data set submodule is used for carrying out empirical mode decomposition on the initial white noise data set to construct a middle white noise data set;
And the target white noise data sub-module is used for carrying out Hilbert transformation on the intermediate white noise data set to construct a target white noise data set.
8. The ac microgrid power metering system according to claim 7, wherein said intermediate white noise data set sub-module comprises:
the decomposing unit is used for decomposing each piece of initial white noise data in the initial white noise data set to obtain a corresponding target component and a target remainder;
the target mean value data unit is used for respectively carrying out mean value operation on the target components and the target residual items corresponding to the initial white noise data of a plurality of the same types to obtain a plurality of corresponding target mean value data, wherein the target mean value data comprises a component mean value and a residual item mean value;
the intermediate white noise data unit is used for carrying out zero-mean normalization on each target mean value data to generate corresponding intermediate white noise data;
an intermediate white noise data set construction unit for constructing an intermediate white noise data set using the plurality of intermediate white noise data.
9. An electronic device comprising a memory and a processor, the memory having stored therein a computer program which, when executed by the processor, causes the processor to perform the steps of the method of metering electrical energy of an ac microgrid according to any one of claims 1 to 5.
10. A computer-readable storage medium, on which a computer program is stored, characterized in that the computer program, when executed, implements the method for metering the electrical energy of an ac microgrid according to any one of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310553477.1A CN116578855A (en) | 2023-05-16 | 2023-05-16 | Electric energy metering method, system, equipment and medium of alternating-current micro-grid |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310553477.1A CN116578855A (en) | 2023-05-16 | 2023-05-16 | Electric energy metering method, system, equipment and medium of alternating-current micro-grid |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116578855A true CN116578855A (en) | 2023-08-11 |
Family
ID=87544888
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310553477.1A Pending CN116578855A (en) | 2023-05-16 | 2023-05-16 | Electric energy metering method, system, equipment and medium of alternating-current micro-grid |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116578855A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117310277A (en) * | 2023-09-25 | 2023-12-29 | 国网四川省电力公司营销服务中心 | Electric energy metering method, system, equipment and medium for off-board charger of electric automobile |
-
2023
- 2023-05-16 CN CN202310553477.1A patent/CN116578855A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117310277A (en) * | 2023-09-25 | 2023-12-29 | 国网四川省电力公司营销服务中心 | Electric energy metering method, system, equipment and medium for off-board charger of electric automobile |
| CN117310277B (en) * | 2023-09-25 | 2024-06-04 | 国网四川省电力公司营销服务中心 | Electric energy metering method, system, equipment and medium for off-board charger of electric automobile |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11349306B2 (en) | Distribution grid fault analysis under load and renewable energy uncertainties | |
| Liu et al. | A Lagrange multiplier based state enumeration reliability assessment for power systems with multiple types of loads and renewable generations | |
| CN110909983B (en) | Multidimensional assessment method for electric energy quality of active power distribution network | |
| Liu et al. | High-precision identification of power quality disturbances under strong noise environment based on FastICA and random forest | |
| Cheng et al. | Enhanced state estimation and bad data identification in active power distribution networks using photovoltaic power forecasting | |
| Ma et al. | Topology identification of distribution networks using a split-EM based data-driven approach | |
| CN103278685A (en) | Harmonic duty quantitative allocation method based on statistical data correlation analysis | |
| CN110222887A (en) | Prediction technique based on VMD and DNN and the application in short-term load forecasting | |
| CN101944776A (en) | Low-frequency oscillation monitoring system of electrical network | |
| CN109470985A (en) | A kind of voltage sag source identification methods based on more resolution singular value decompositions | |
| Kaddah et al. | Probabilistic power quality indices for electric grids with increased penetration level of wind power generation | |
| CN103020459A (en) | Method and system for sensing multiple-dimension electric utilization activities | |
| CN116578855A (en) | Electric energy metering method, system, equipment and medium of alternating-current micro-grid | |
| CN110244185A (en) | A kind of multi-source harmonic contributions division methods, terminal device and storage medium | |
| Xu et al. | Interval prediction method for wind power based on VMD-ELM/ARIMA-ADKDE | |
| CN115983507B (en) | Method and system for predicting broadband oscillation risk of section of power grid of transmitting end source | |
| CN116298509A (en) | Online identification method for harmonic resonance of power system | |
| Coury et al. | Programmable logic design of a compact Genetic Algorithm for phasor estimation in real-time | |
| Cho et al. | Oscillation recognition using a geometric feature extraction process based on periodic time-series approximation | |
| Jose et al. | Automatic and real time classification of power quality disturbance using statistical moments | |
| Wang et al. | Short term load forecasting based on vmd-dnn | |
| Behzadi et al. | Dual identification of multi-complex and non-stationary power quality disturbances using variational mode decomposition in hybrid modern power systems | |
| Igual et al. | Evaluation of automatic power quality classification in microgrids operating in islanded mode | |
| Yue et al. | A power system harmonic problem based on the BP neural network learning algorithm | |
| Jiandong et al. | Daily electricity consumption combined forecast based on variational mode decomposition and sample entropy |
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 |