CN117096956A - Harmonic control method and system of high-voltage frequency converter - Google Patents
Harmonic control method and system of high-voltage frequency converter Download PDFInfo
- Publication number
- CN117096956A CN117096956A CN202311362543.3A CN202311362543A CN117096956A CN 117096956 A CN117096956 A CN 117096956A CN 202311362543 A CN202311362543 A CN 202311362543A CN 117096956 A CN117096956 A CN 117096956A
- Authority
- CN
- China
- Prior art keywords
- harmonic
- groups
- data sets
- filtering
- module
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 46
- 238000001914 filtration Methods 0.000 claims abstract description 239
- 238000004458 analytical method Methods 0.000 claims abstract description 96
- 238000012545 processing Methods 0.000 claims abstract description 8
- 238000012544 monitoring process Methods 0.000 claims description 76
- 230000004913 activation Effects 0.000 claims description 69
- 238000012549 training Methods 0.000 claims description 42
- 238000013507 mapping Methods 0.000 claims description 22
- 238000010276 construction Methods 0.000 claims description 16
- 238000004364 calculation method Methods 0.000 claims description 14
- 230000002776 aggregation Effects 0.000 claims description 9
- 238000004220 aggregation Methods 0.000 claims description 9
- 230000002452 interceptive effect Effects 0.000 claims description 8
- 230000001360 synchronised effect Effects 0.000 claims description 5
- 238000005259 measurement Methods 0.000 claims description 3
- 238000010183 spectrum analysis Methods 0.000 claims description 3
- 230000000694 effects Effects 0.000 abstract description 8
- 230000006872 improvement Effects 0.000 abstract description 4
- 238000010586 diagram Methods 0.000 description 6
- 230000005540 biological transmission Effects 0.000 description 4
- 230000001939 inductive effect Effects 0.000 description 4
- 230000003044 adaptive effect Effects 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000012795 verification Methods 0.000 description 3
- 230000009467 reduction Effects 0.000 description 2
- 238000012163 sequencing technique Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 108010025037 T140 peptide Proteins 0.000 description 1
- 230000002411 adverse Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000013506 data mapping Methods 0.000 description 1
- 230000003090 exacerbative effect Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 230000015654 memory Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000013021 overheating Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
- H02J3/01—Arrangements for reducing harmonics or ripples
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
- H02J3/38—Arrangements for parallely feeding a single network by two or more generators, converters or transformers
- H02J3/46—Controlling of the sharing of output between the generators, converters, or transformers
-
- 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
Abstract
The invention provides a harmonic control method and a system of a high-voltage frequency converter, which relate to the technical field of data processing, and pass filtered data sets and deviation filtered data sets are obtained by traversing history filtered data sets based on harmonic constraint parameters, and conventional harmonic data sets and deviation harmonic data sets are obtained; constructing a first-level harmonic control analysis sub-network based on the qualified filtering data set and the conventional harmonic data set; traversing the qualified filtering dataset and the conventional harmonic dataset by taking the deviation harmonic dataset as constraint to obtain a fitting filtering dataset and constructing a secondary harmonic control analysis sub-network; and synchronizing the two harmonic control analysis sub-networks to the intelligent filtering module for harmonic automatic control. Solves the technical problems of insufficient effectiveness and stability of harmonic filtering in the prior art. The technical effects of multistage adjustment of the filtering parameters according to the real-time harmonic parameters, improvement of the effectiveness and stability of harmonic filtering and guarantee of the operation safety of the circuit are achieved.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to a harmonic control method and system of a high-voltage frequency converter.
Background
In current power systems, there are drawbacks of insufficient effectiveness and stability of harmonic filtering, which may raise a series of problems including the risk of inducing failure and damage to the powered device.
When harmonics are not effectively filtered, they may propagate through the power system into the powered device, causing current distortion and voltage fluctuations, and such non-standard power waveforms may adversely affect various electrical devices, including but not limited to exacerbating thermal losses of the device, causing overheating, reducing the reliability and lifetime of the device.
In summary, the prior art has the technical problems of insufficient effectiveness and stability of harmonic filtering and risk of inducing faults and damages of electric equipment.
Disclosure of Invention
The application provides a harmonic control method and a harmonic control system for a high-voltage frequency converter, which are used for solving the technical problems that in the prior art, the effectiveness and stability of harmonic filtering are insufficient, and the risk of inducing the faults and damage of electric equipment exists.
In view of the above problems, the present application provides a method and a system for controlling harmonics of a high-voltage inverter.
In a first aspect of the present application, there is provided a harmonic control method of a high voltage frequency converter, the method comprising: acquiring historical harmonic output information, wherein the historical harmonic output information is acquired through a first monitoring module of an interactive N-group harmonic control circuit, and the historical harmonic output information comprises N groups of historical harmonic data sets; acquiring historical filtering output information, wherein the historical filtering output information is acquired through a second monitoring module interacting with the N groups of harmonic control lines, and the historical filtering output information comprises N groups of historical filtering data sets; presetting harmonic constraint parameters, traversing the N groups of historical filtering data sets based on the harmonic constraint parameters, and obtaining N groups of grid filtering data sets and N groups of deviation filtering data sets; mapping and dividing the N groups of historical harmonic data sets according to the N groups of grid filter data sets and the N groups of deviation filter data sets to obtain N groups of conventional harmonic data sets and N groups of deviation harmonic data sets; constructing a first-order harmonic control analysis sub-network based on federal learning, and training the first-order harmonic control analysis sub-network based on the N-group conventional harmonic data sets and the N-group filtered data sets; traversing the N groups of conventional harmonic data sets and the N groups of deviation filtering data sets by taking the N groups of deviation filtering data sets as constraints to obtain N groups of fitting filtering data sets; pre-constructing a secondary harmonic control analysis sub-network, and training the secondary harmonic control analysis sub-network based on the N groups of deviation filtering data sets and the N groups of fitting filtering data sets; and synchronizing the first-level harmonic control analysis sub-network and the second-level harmonic control analysis sub-network to the intelligent filtering module of the N groups of harmonic control lines, and performing harmonic automatic control based on the intelligent filtering module.
In a second aspect of the present application, there is provided a harmonic control system for a high voltage frequency converter, the system comprising: the system comprises a harmonic information acquisition unit, a first monitoring module and a second monitoring module, wherein the harmonic information acquisition unit is used for acquiring historical harmonic output information, the historical harmonic output information is acquired through the first monitoring module of an interactive N-group harmonic control circuit, and the historical harmonic output information comprises N groups of historical harmonic data sets; the filtering information acquisition unit is used for acquiring historical filtering output information, wherein the historical filtering output information is acquired through a second monitoring module which interacts with the N groups of harmonic control lines, and the historical filtering output information comprises N groups of historical filtering data sets; the harmonic constraint setting unit is used for presetting harmonic constraint parameters, traversing the N groups of historical filtering data sets based on the harmonic constraint parameters, and obtaining N groups of grid filtering data sets and N groups of deviation filtering data sets; the data division execution unit is used for mapping and dividing the N groups of historical harmonic data sets according to the N groups of grid filter data sets and the N groups of deviation filter data sets to obtain N groups of conventional harmonic data sets and N groups of deviation harmonic data sets; the harmonic analysis construction unit is used for constructing a first-order harmonic control analysis sub-network based on federal learning and training the first-order harmonic control analysis sub-network based on the N-group conventional harmonic data sets and the N-group combination lattice filter data sets; the data filtering execution unit is used for traversing the N groups of conventional harmonic data sets and the N groups of deviation filtering data sets by taking the N groups of deviation filtering data sets as constraints to obtain N groups of fitting filtering data sets; the second-level control construction unit is used for pre-constructing a second-level harmonic control analysis sub-network and training the second-level harmonic control analysis sub-network based on the N groups of deviation filtering data sets and the N groups of fitting filtering data sets; and the harmonic correlation control unit is used for synchronizing the first-level harmonic control analysis sub-network and the second-level harmonic control analysis sub-network to the intelligent filtering modules of the N groups of harmonic control circuits, and carrying out harmonic automatic control based on the intelligent filtering modules.
One or more technical schemes provided by the application have at least the following technical effects or advantages:
the method provided by the embodiment of the application comprises the steps of obtaining historical harmonic output information, wherein the historical harmonic output information is obtained through a first monitoring module of an interactive N-group harmonic control circuit, and the historical harmonic output information comprises N groups of historical harmonic data sets; acquiring historical filtering output information, wherein the historical filtering output information is acquired through a second monitoring module interacting with the N groups of harmonic control lines, and the historical filtering output information comprises N groups of historical filtering data sets; presetting harmonic constraint parameters, traversing the N groups of historical filtering data sets based on the harmonic constraint parameters, and obtaining N groups of grid filtering data sets and N groups of deviation filtering data sets; mapping and dividing the N groups of historical harmonic data sets according to the N groups of grid filter data sets and the N groups of deviation filter data sets to obtain N groups of conventional harmonic data sets and N groups of deviation harmonic data sets; constructing a first-order harmonic control analysis sub-network based on federal learning, and training the first-order harmonic control analysis sub-network based on the N-group conventional harmonic data sets and the N-group filtered data sets; traversing the N groups of deviation harmonic data sets and the N groups of conventional harmonic data sets by taking the N groups of deviation harmonic data sets as constraints to obtain N groups of fitting filter data sets; pre-constructing a secondary harmonic control analysis sub-network, and training the secondary harmonic control analysis sub-network based on the N groups of deviation harmonic data sets and the N groups of fitting filtering data sets; and synchronizing the first-level harmonic control analysis sub-network and the second-level harmonic control analysis sub-network to the intelligent filtering module of the N groups of harmonic control lines, and performing harmonic automatic control based on the intelligent filtering module. The technical effects of multistage adjustment of filtering parameters according to real-time harmonic parameters, improvement of harmonic filtering effectiveness and stability, guarantee of circuit operation safety and reduction of failure damage risk of electric equipment are achieved.
Drawings
Fig. 1 is a schematic flow chart of a harmonic control method of a high-voltage frequency converter provided by the application;
fig. 2 is a schematic flow chart of determining real-time harmonic parameters in a harmonic control method of a high-voltage frequency converter according to the present application;
fig. 3 is a schematic structural diagram of a harmonic control device of a high-voltage frequency converter according to the present application;
FIG. 4 is a schematic diagram of a conventional harmonic control apparatus according to the present application;
FIG. 5 is a schematic flow chart of constructing a first harmonic control analysis sub-network in the harmonic control method of the high-voltage frequency converter provided by the application;
fig. 6 is a schematic structural diagram of a harmonic control system of a high-voltage inverter according to the present application.
Reference numerals illustrate: the device comprises a harmonic information acquisition unit 1, a filtering information acquisition unit 2, a harmonic constraint setting unit 3, a data dividing and executing unit 4, a harmonic analysis construction unit 5, a data filtering and executing unit 6, a secondary control construction unit 7 and a harmonic association control unit 8.
Detailed Description
The application provides a harmonic control method and a harmonic control system for a high-voltage frequency converter, which are used for solving the technical problems that in the prior art, the effectiveness and stability of harmonic filtering are insufficient, and the risk of inducing the faults and damage of electric equipment exists. The technical effects of multistage adjustment of filtering parameters according to real-time harmonic parameters, improvement of harmonic filtering effectiveness and stability, guarantee of circuit operation safety and reduction of failure damage risk of electric equipment are achieved.
The technical scheme of the application accords with related regulations on data acquisition, storage, use, processing and the like.
In the following, the technical solutions of the present application will be clearly and completely described with reference to the accompanying drawings, and it should be understood that the described embodiments are only some embodiments of the present application, but not all embodiments of the present application, and that the present application is not limited by the exemplary embodiments described herein. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application. It should be further noted that, for convenience of description, only some, but not all of the drawings related to the present application are shown.
Example 1
As shown in fig. 1, the present application provides a harmonic control method of a high-voltage frequency converter, the method is applied to a harmonic control system of the high-voltage frequency converter, the system includes N groups of harmonic control lines, each group of harmonic control lines includes a first monitoring module, a second monitoring module and an intelligent filtering module, and the method includes:
t100, acquiring historical harmonic output information, wherein the historical harmonic output information is acquired through a first monitoring module of an interactive N-group harmonic control circuit, and the historical harmonic output information comprises N groups of historical harmonic data sets;
In one embodiment, as shown in fig. 2, the method steps provided by the present application further include:
t110, the first monitoring module comprises a current signal acquisition sub-module, a voltage signal acquisition sub-module, a harmonic component analysis sub-module and a harmonic parameter calculation sub-module;
t120, obtaining a real-time current signal based on the current signal acquisition sub-module and obtaining a real-time voltage signal based on the voltage signal acquisition sub-module;
t130, the harmonic component analysis submodule performs waveform measurement and spectrum analysis based on the real-time current signal and the real-time voltage signal to obtain a real-time harmonic component;
and T140, the harmonic parameter calculation sub-module determines real-time harmonic parameters according to the real-time harmonic components.
Specifically, in this embodiment, a harmonic control method of a high-voltage inverter is applied to a harmonic control apparatus of a high-voltage inverter, and a schematic structural diagram of the apparatus is shown in fig. 3, in the system, a target inverter has N groups (phase) of harmonic control lines, and components for harmonic control in each group of harmonic control lines are identical in composition and each group of harmonic control lines is composed of a first monitoring module, a second monitoring module and an intelligent filtering module, where n=3 generally.
The intelligent filtering module is specifically composed of a primary filtering module and a secondary filtering module which are in communication connection, the secondary filtering module selectively activates and operates according to the filtering effect of the primary filtering module on the harmonic wave generated by the frequency converter, and the harmonic wave transmitted to a power grid line in the operation process of the target frequency converter is performed based on the intelligent filtering module.
The arrangement condition of each module in each group of harmonic control lines is a first monitoring module, a first-stage filtering module, a second monitoring module and a second-stage filtering module which are sequentially connected, and the functions of the first monitoring module and the second monitoring module are consistent, and the functions of the first monitoring module and the second monitoring module are further analyzed by obtaining real-time current signals and real-time voltage signals of the harmonic control lines, so that real-time harmonic parameters are obtained and data are recorded.
The method of obtaining and recording real-time harmonic parameters by the first monitoring module is taken as an example to carry out detailed explanation of the technical scheme.
Specifically, the first monitoring module is specifically composed of a current signal acquisition sub-module, a voltage signal acquisition sub-module, a harmonic component analysis sub-module and a harmonic parameter calculation sub-module.
And based on the current signal acquisition sub-module and the voltage signal acquisition sub-module, acquiring real-time current signals and real-time voltage signals of the corresponding harmonic control circuit to obtain real-time current signals and real-time voltage signals.
The real-time current signal and the real-time voltage signal are input into the harmonic component analysis submodule, and a Fast Fourier Transform (FFT) is adopted in the harmonic component analysis submodule to convert a time domain signal into a frequency domain signal, wherein the frequency domain signal is the real-time harmonic component. Further, the harmonic parameter calculation sub-module analyzes the amplitude and the phase of each frequency component according to the real-time harmonic component to obtain a real-time harmonic parameter, wherein the real-time harmonic component specifically comprises the frequency, the amplitude and the phase.
As shown in fig. 4, which is a schematic structural diagram of a conventional high-voltage inverter harmonic control device, each group of harmonic control circuits of a conventional N groups (phase) of harmonic control circuits includes a first monitoring module, a second monitoring module and a first-stage filtering module, wherein the second monitoring module is used for monitoring a current voltage after the target inverter output harmonic filtering through the first-stage filtering module, analyzing and obtaining a harmonic condition of an actual circuit after the harmonic filtering through the first-stage filtering module, and in the conventional high-voltage inverter harmonic control system, the intelligent filtering module only includes the first-stage filtering module.
Thus, the present embodiment is based on obtaining, by the first monitoring module of the interactive N sets of harmonic control lines, the historical harmonic output information of the historical target frequency converter input N-phase harmonic control lines, the historical harmonic output information correspondingly comprising N sets of historical harmonic data sets.
T200, acquiring historical filtering output information, wherein the historical filtering output information is acquired through a second monitoring module interacting with the N groups of harmonic control lines, and the historical filtering output information comprises N groups of historical filtering data sets;
specifically, in this embodiment, the second monitoring modules of the N groups of harmonic control circuits are interacted to obtain the history filtering output information, where the history filtering output information includes N groups of history filtering data sets, and the history filtering output information is the harmonic condition in the circuit after the N primary filtering modules perform harmonic filtering.
Meanwhile, it should be understood that there is a one-to-one mapping relationship between the N sets of historical filtered data sets and the N sets of historical harmonic data sets based on line groups and time.
T300, presetting harmonic constraint parameters, traversing the N groups of historical filtering data sets based on the harmonic constraint parameters, and obtaining N groups of combined lattice filtering data sets and N groups of deviation filtering data sets;
T400, mapping and dividing the N groups of historical harmonic data sets according to the N groups of grid filter data sets and the N groups of deviation filter data sets to obtain N groups of conventional harmonic data sets and N groups of deviation harmonic data sets;
specifically, in this embodiment, the preset harmonic constraint parameter is a maximum harmonic parameter allowed by the power transmission line under the condition that the N-phase power transmission line operates stably. The preset harmonic constraint parameters are set according to the specific use situation of the power transmission line, the numerical values of the preset harmonic constraint parameters are not limited forcedly in the embodiment, and the preset harmonic constraint parameters are specifically composed of frequency constraint, amplitude constraint and phase constraint.
In this embodiment, the N sets of history filter datasets are traversed based on the harmonic constraint parameter to divide each set of history filter datasets into a qualified filter dataset that is qualified for harmonic filtering and a deviation filter dataset that is unqualified for harmonic filtering.
And further carrying out mapping division on the N groups of historical harmonic data sets according to the N groups of historical harmonic data sets and the N groups of deviation filtering data sets based on the one-to-one mapping relation between the line groups and the time to obtain N groups of conventional harmonic data sets and N groups of deviation harmonic data sets.
The N groups of conventional harmonic data sets and the N groups of grid-combined filter data sets are mapped in time, and the N groups of deviation harmonic data sets and the N groups of deviation filter data sets are mapped in time, namely when the harmonic wave output by the target frequency converter falls into the N groups of conventional harmonic data sets, the first-stage filter module can perform effective harmonic filtering, otherwise, when the harmonic wave output by the target frequency converter falls into the N groups of deviation harmonic data sets, the first-stage filter module cannot perform effective harmonic filtering, and the harmonic wave affecting the running reliability of the line still exists.
Based on this, the conventional harmonic filtering device as shown in fig. 3 is improved, and a harmonic control system with a newly added secondary filtering module for assisting in harmonic secondary filtering is obtained.
The embodiment obtains the N groups of conventional harmonic data sets and N groups of lattice filter data sets, where the N groups of deviation harmonic data sets and N groups of deviation filter data sets are used for implementing intelligent adjustment of filter control parameters of the primary filter module by using the N groups of deviation harmonic data sets and N groups of deviation filter data sets as analysis data.
T500, constructing a first-order harmonic control analysis sub-network based on federal learning, and training the first-order harmonic control analysis sub-network based on the N-group conventional harmonic data sets and the N-group filtered data sets;
In one embodiment, as shown in fig. 5, a first harmonic control analysis sub-network is constructed based on federal learning, and training of the first harmonic control analysis sub-network is performed based on the N-combination lattice filter dataset and the N-group conventional harmonic dataset, and a method step T500 provided by the present application further includes:
t510, the intelligent filtering module comprises a primary filtering module and a secondary filtering module;
t520, acquiring a historical harmonic control parameter set, wherein the historical harmonic control parameter set is acquired by an intelligent filtering module interacting the N groups of harmonic control lines, and the historical harmonic control parameter set comprises N groups of historical harmonic control parameter information;
t530, carrying out mapping call on the N groups of historical harmonic control parameter information according to the N combination lattice filter data sets to obtain N combination lattice control parameter sets;
t540, pre-building a standard harmonic controller, wherein the standard harmonic controller comprises a decoder and an encoder;
t550, obtaining N harmonic controllers, wherein the N harmonic controllers are obtained by performing supervision training on the standard harmonic controller by adopting the N combined lattice filter data sets, the N groups of conventional harmonic data sets and the N combined lattice control parameter sets;
T560, extracting controller parameters of the N harmonic controllers to obtain N groups of controller parameters;
t570, performing aggregation processing on the N groups of controller parameters based on a federal aggregation algorithm to obtain optimized controller parameters;
t580, adopting the optimized controller parameters to update the parameters of the standard harmonic controller to obtain the primary harmonic control analysis sub-network;
t590, N primary filtering modules of the N groups of harmonic control lines to which the primary harmonic control analysis sub-network is synchronized.
In one embodiment, N harmonic controllers are obtained, where the N harmonic controllers are obtained by performing supervised training of the standard harmonic controller using the N-combination lattice filtered data set and the N-group conventional harmonic data set, and the method step T550 provided by the present application further includes:
t551, performing mapping call on the N-combination lattice filter data set, the N groups of conventional harmonic data sets and the N-combination lattice control parameter sets to obtain a first qualified filter data set, a first conventional harmonic data set and a first qualified control parameter set;
t552, performing supervised training of a decoder and an encoder of the standard harmonic controller by adopting the first qualified filtering data set, the first conventional harmonic data set and the first qualified control parameter set to obtain a first harmonic controller;
And T553, performing supervision training on the standard harmonic controller by adopting the N-combination lattice filter data set and the N groups of conventional harmonic data sets to obtain the N harmonic controllers.
Specifically, in this embodiment, the intelligent filtering module includes a primary filtering module and a secondary filtering module, and a second monitoring module is disposed between the primary filtering module and the secondary filtering module.
Because of the non-uniformity of the harmonic control parameter settings for the harmonic filtering of the different harmonic intelligent filtering modules, and based on step S100, in the conventional high-voltage inverter harmonic control system, the intelligent filtering module only includes one stage of filtering module, so that the embodiment obtains the historical harmonic control parameter set by interacting the intelligent filtering modules (i.e., the one stage of filtering modules) of the N sets of harmonic control lines as shown in fig. 4, where the historical harmonic control parameter set includes N sets of historical harmonic control parameter information.
And then, based on the N groups of historical filtering data sets, the N groups of historical harmonic data sets and the N groups of historical harmonic control parameter information, a one-to-one mapping relation exists necessarily based on the line groups and time, mapping and calling are carried out on the N groups of historical harmonic control parameter information according to the N groups of historical harmonic control parameter information to obtain N groups of historical harmonic control parameter sets, wherein the N groups of historical harmonic control parameter sets are the harmonic control parameter information when the first-stage filtering module effectively filters the harmonic output by the target frequency converter.
According to the method, the device and the system, the harmonic data, the filtering data and the harmonic control data in an effective harmonic filtering scene are screened, so that the technical effect of improving the effectiveness and accuracy of the harmonic control data output by the sub-network is provided for the construction of the sub-network for carrying out the primary harmonic control analysis for carrying out harmonic control parameter setting analysis according to real-time harmonic.
In the present embodiment, N harmonic controllers for performing harmonic control parameter analysis of only the corresponding harmonic control line are first constructed, and the structures of the N harmonic controllers are identical. The present embodiment thus pre-constructs a standard harmonic controller, wherein the standard harmonic controller includes a decoder and an encoder.
The standard harmonic controllers are adopted to call different data for output precision training so as to obtain N harmonic controllers, and because all the N harmonic controllers are obtained by model supervision training by taking the standard harmonic controllers as basic model structures, the output precision training method of the N harmonic controllers has consistency, and based on the fact, the embodiment takes building one harmonic controller as an example for carrying out detailed explanation of a technical scheme.
Specifically, in this embodiment, the first qualified filtered data set, the first conventional harmonic data set, and the first qualified control parameter set mapped with the random first harmonic control line of the N sets of harmonic control lines are obtained by performing a mapping call on the N sets of conventional harmonic data sets, and the N sets of control parameters.
Dividing the first qualified filter data set, the first conventional harmonic data set and the first qualified control parameter set identification into a training set and a testing set by adopting a data volume of 9:1, performing supervised training of a decoder and an encoder of the standard harmonic controller based on the training set and the testing set, extracting random data from the first qualified filter data set, the first conventional harmonic data set and the first qualified control parameter set to obtain a verification set, and performing output precision verification of the decoder and the encoder of the standard harmonic controller based on the verification set until the output precision is stable to be higher than 97%, and outputting the first harmonic controller. And similarly, performing supervision training of the standard harmonic controller by adopting the N-combination lattice filter data set and the N groups of conventional harmonic data sets to obtain the N harmonic controllers.
And extracting controller parameters, namely model parameters, of the N harmonic controllers to obtain N groups of controller parameters, wherein parameter index items of each group of controller parameters have consistency. The federation aggregation algorithm in this embodiment is mean calculation of the same-parameter index items, and the parameter extraction aggregation and mean calculation of the same-parameter index items of the N groups of controller parameters are performed based on the federation aggregation algorithm to obtain the optimized controller parameters.
And transmitting the parameters of the optimized controller to the parameter update of the standard harmonic controller to obtain the primary harmonic control analysis sub-network, wherein the primary harmonic control analysis sub-network can adapt to the harmonic control requirements of N groups of harmonic control lines.
And N primary filtering modules of the N groups of harmonic control circuits to which the primary harmonic control analysis sub-network is synchronized are used for effectively filtering the harmonic wave within a certain harmonic parameter range based on the primary filtering modules.
According to the embodiment, the first-stage filtering module is constructed based on federal learning, so that the intelligent harmonic controller meeting the harmonic filtering requirements of harmonic control lines of different phases is obtained, and the technical effect of improving the filtering effectiveness of the harmonic generated by the frequency converter is achieved.
T600, traversing the N groups of the deviation filtering data sets and the N groups of the conventional harmonic data sets by taking the N groups of the deviation filtering data sets as constraints to obtain N groups of fitting filtering data sets;
in particular, it should be appreciated that the N sets of bias filter datasets are "legacy harmonics" that were not successfully filtered by the primary filter module, and thus the present embodiment configures the secondary harmonic filter module to enable secondary filtering of the "legacy harmonics".
According to the embodiment, the N groups of conventional harmonic data sets are traversed by taking the N groups of deviation filtering data sets as references to obtain conventional harmonics which are consistent with the 'legacy harmonic' values of the N groups of deviation filtering data sets and fall into the N groups of conventional harmonic data sets, and then corresponding data call is conducted on the N groups of lattice filtering data sets based on the data time information of the obtained conventional harmonics, so that the N groups of fitting filtering data sets are obtained.
When the N groups of deviation filtering data sets are the harmonic waves transmitted by the target frequency converter, the theoretical filtering result is the N groups of fitting filtering data sets after harmonic wave filtering is carried out by the first-stage filtering module.
T700, pre-constructing a secondary harmonic control analysis sub-network, and training the secondary harmonic control analysis sub-network based on the N groups of deviation filtering data sets and the N groups of fitting filtering data sets;
In one embodiment, the method steps provided by the application further comprise:
t711, carrying out serialization processing on the N groups of deviation harmonic data sets to obtain deviation harmonic extremum;
t712, the first monitoring module further comprises an activation judgment execution sub-module;
t713, constructing an activation judgment parameter based on the deviation harmonic extremum;
t714, synchronizing the activation judgment parameters to the activation judgment execution sub-module;
t715, carrying out numerical judgment on the real-time harmonic parameter and the deviation harmonic extremum based on the activation judgment execution submodule;
and T716, when the real-time harmonic parameter is larger than the deviation harmonic extremum, sending an activation instruction to activate the secondary filtering module.
In one embodiment, the method further comprises the steps of pre-constructing a second harmonic control analysis sub-network, and training the second harmonic control analysis sub-network based on the N sets of deviation filtering data sets and the N sets of fitting filtering data sets:
t721, performing mapping call on the N groups of historical harmonic control parameter information according to the N groups of fitting filtering data sets to obtain N groups of fitting control parameter sets;
and T722, performing supervision training of the standard harmonic controller by adopting the N groups of deviation filtering data sets, the N groups of fitting filtering data sets and the N groups of fitting control parameter sets to obtain the secondary harmonic control analysis sub-network.
Specifically, in this embodiment, the second harmonic control analysis sub-network and the first harmonic control analysis sub-network have the same functions, and may perform harmonic control parameter generation of the first filtering module/the second filtering module according to the harmonic parameters, so as to perform effective harmonic filtering.
The construction method of the secondary harmonic control analysis subnetwork comprises the following steps:
carrying out data mapping calling of the same-time nodes on the N groups of historical harmonic control parameter information according to the data time information of the N groups of fitting filtering data sets to obtain N groups of fitting control parameter sets;
and taking the N groups of deviation filtering data sets, the N groups of fitting filtering data sets and the N groups of fitting control parameter sets as training data, and performing supervision training of the standard harmonic controller by adopting the same method of training any N harmonic controllers to obtain the secondary harmonic control analysis subnetwork.
And synchronizing the secondary harmonic control analysis sub-network to N secondary filtering modules of the N groups of harmonic control circuits so as to realize the secondary filtering of the harmonic which is not successfully filtered by the primary filtering module based on the secondary filtering modules, thereby effectively reducing the harmonic.
The activation conditions of the secondary filtering module are as follows:
and sequencing the N groups of deviation harmonic data sets from large to small to obtain N minimum deviation harmonic data, sequencing the N minimum deviation harmonic data from large to small to obtain the deviation harmonic extremum of the minimum harmonic data represented in the N groups of deviation harmonic data sets, and constructing an activation judgment parameter based on the deviation harmonic extremum.
The first monitoring module further comprises an activation judgment execution sub-module, and the activation judgment execution sub-module synchronizes the activation judgment parameters.
When the first monitoring module monitors a real-time current signal and a real-time voltage signal and performs harmonic analysis to obtain a real-time harmonic parameter, the numerical judgment of the real-time harmonic parameter and the deviation harmonic extremum is performed based on the activation judgment execution submodule.
When the real-time harmonic parameter is larger than the deviation harmonic extremum, the first monitoring module sends an activation instruction to the secondary filtering activation module, the activation instruction is transmitted to the activation control execution module based on the secondary filtering activation module, and the activation control execution module issues the activation instruction to the secondary filtering module belonging to the same harmonic control circuit for module activation.
As shown in fig. 3, a schematic structure diagram of a harmonic control device of a high-voltage frequency converter is provided, and a secondary filtering activation module and an activation control execution module are added in the implementation, so that unified management of activation instructions is realized, and effectiveness of transmission of the activation instructions is improved.
And T800, synchronizing the first-level harmonic control analysis sub-network and the second-level harmonic control analysis sub-network to the intelligent filtering module of the N groups of harmonic control lines, and carrying out harmonic automatic control based on the intelligent filtering module.
In one embodiment, the method steps provided by the application further comprise:
t810, constructing a harmonic control monitoring window;
t820, performing activation frequency statistics of the secondary filter module based on the harmonic control monitoring window to obtain a plurality of secondary activation frequency information, wherein the plurality of secondary activation frequency information corresponds to a plurality of harmonic control monitoring windows;
t830, carrying out weight assignment on the secondary activation frequency information according to the time span of the harmonic control monitoring windows and the current time, and obtaining steady activation frequency information through weight calculation;
t840, presetting a filter module replacement threshold, and judging whether the steady-state activation frequency information meets the filter module replacement threshold;
And T850, if the steady-state activation frequency information meets the filter module replacement threshold, generating a filter module replacement instruction.
Specifically, in this embodiment, the first harmonic control analysis sub-network and the second harmonic control analysis sub-network are synchronized to the first filtering module and the second filtering module of the intelligent filtering module of the N groups of harmonic control lines, harmonic parameter generation is performed based on the first monitoring module and the second monitoring module, and the first filtering module and the second filtering module perform corresponding filtering processing according to the harmonic parameters sent by the first monitoring module and the second monitoring module, so as to implement automatic harmonic control based on the intelligent filtering module.
Further, as shown in fig. 4, the present embodiment constructs a harmonic control monitoring window, where the harmonic control monitoring window is configured in the harmonic control monitoring module of fig. 4, and an exemplary time interval of each of the harmonic control monitoring windows is 15min.
And counting the activation frequencies of the secondary filter modules based on the harmonic control monitoring windows to obtain a plurality of secondary activation frequency information, wherein the plurality of secondary activation frequency information corresponds to the plurality of harmonic control monitoring windows, and each secondary activation frequency information is the total activation frequency of N secondary filter modules in the time span of the harmonic control monitoring windows.
And carrying out weight assignment on the plurality of second-level activation frequency information according to the time span of the plurality of harmonic control monitoring windows and the current time so as to realize that the weight of the harmonic monitoring window which is closer to the current time is higher, and carrying out weight calculation on the basis of the weight assignment result and the plurality of second-level activation frequency information to obtain steady-state activation frequency information, wherein the steady-state activation frequency information is a prediction result of N second-level filter module activation frequencies in the current time and the subsequent harmonic control monitoring window.
The filter module replacement threshold value used for evaluating whether the first-stage filter module is the adaptive harmonic control circuit is preset, whether the steady-state activation frequency information meets the filter module replacement threshold value is judged, if the steady-state activation frequency information meets the filter module replacement threshold value, the fact that the filtering performance of the current first-stage filter module is not adaptive to the harmonic filtering requirement of the harmonic control circuit is indicated, and therefore a filter module replacement instruction is generated to prompt operation and maintenance personnel to replace the filter model of the first-stage filter module.
According to the embodiment, the adaptive situation of the primary filtering module is timely and effectively obtained by adding the harmonic control monitoring window and the harmonic control monitoring module, so that the technical effect of timely replacing a proper filter and improving the operation safety and stability of a harmonic control circuit is achieved.
Example two
Based on the same inventive concept as the harmonic control method of a high-voltage inverter in the foregoing embodiments, as shown in fig. 6, the present application provides a harmonic control system of a high-voltage inverter, wherein the system includes:
the harmonic information acquisition unit 1 is used for acquiring historical harmonic output information, wherein the historical harmonic output information is acquired through a first monitoring module of an interactive N-group harmonic control circuit, and the historical harmonic output information comprises N groups of historical harmonic data sets;
the filtering information acquisition unit 2 is used for acquiring historical filtering output information, wherein the historical filtering output information is acquired through a second monitoring module which interacts with the N groups of harmonic control circuits, and the historical filtering output information comprises N groups of historical filtering data sets;
the harmonic constraint setting unit 3 is used for presetting harmonic constraint parameters and traversing the N groups of historical filtering data sets based on the harmonic constraint parameters to obtain N groups of grid filtering data sets and N groups of deviation filtering data sets;
the data division execution unit 4 is used for mapping and dividing the N groups of historical harmonic data sets according to the N groups of grid filter data sets and the N groups of deviation filter data sets to obtain N groups of conventional harmonic data sets and N groups of deviation harmonic data sets;
The harmonic analysis construction unit 5 is used for constructing a first-order harmonic control analysis sub-network based on federal learning and training the first-order harmonic control analysis sub-network based on the N-group combined lattice filtering data set and the N-group conventional harmonic data set;
the data filtering execution unit 6 is used for traversing the N groups of the conventional harmonic data sets and the N groups of the deviation filtering data sets by taking the N groups of the deviation filtering data sets as constraints to obtain N groups of fitting filtering data sets;
the secondary control construction unit 7 is used for pre-constructing a secondary harmonic control analysis sub-network and training the secondary harmonic control analysis sub-network based on the N groups of deviation filtering data sets and the N groups of fitting filtering data sets;
and the harmonic correlation control unit 8 is used for synchronizing the primary harmonic control analysis sub-network and the secondary harmonic control analysis sub-network to the intelligent filter modules of the N groups of harmonic control circuits, and carrying out harmonic automatic control based on the intelligent filter modules.
In one embodiment, the harmonic analysis construction unit 5 further comprises:
the intelligent filtering module comprises a primary filtering module and a secondary filtering module;
acquiring a historical harmonic control parameter set, wherein the historical harmonic control parameter set is obtained by an intelligent filtering module interacting the N groups of harmonic control lines, and the historical harmonic control parameter set comprises N groups of historical harmonic control parameter information;
Performing mapping call on the N groups of historical harmonic control parameter information according to the N combination lattice filter data sets to obtain N combination lattice control parameter sets;
pre-constructing a standard harmonic controller, wherein the standard harmonic controller comprises a decoder and an encoder;
obtaining N harmonic controllers, wherein the N harmonic controllers are obtained by performing supervised training of the standard harmonic controller by adopting the N combined lattice filter data sets, the N groups of conventional harmonic data sets and the N combined lattice control parameter sets;
extracting controller parameters of the N harmonic controllers to obtain N groups of controller parameters;
performing aggregation treatment on the N groups of controller parameters based on a federal aggregation algorithm to obtain optimized controller parameters;
the parameters of the standard harmonic controller are updated by adopting the parameters of the optimized controller, and the primary harmonic control analysis sub-network is obtained;
and N primary filtering modules of the N groups of harmonic control circuits to which the primary harmonic control analysis sub-network is synchronized.
In one embodiment, the harmonic analysis construction unit 5 further comprises:
performing mapping call on the N-combination lattice filter data set, the N groups of conventional harmonic data sets and the N-combination lattice control parameter sets to obtain a first qualified filter data set, a first conventional harmonic data set and a first qualified control parameter set;
Performing supervised training of a decoder and an encoder of the standard harmonic controller by adopting the first qualified filtering data set, the first conventional harmonic data set and the first qualified control parameter set to obtain a first harmonic controller;
and similarly, performing supervision training of the standard harmonic controller by adopting the N-combination lattice filter data set and the N groups of conventional harmonic data sets to obtain the N harmonic controllers.
In one embodiment, the harmonic analysis construction unit 5 further comprises:
the first monitoring module comprises a current signal acquisition sub-module, a voltage signal acquisition sub-module, a harmonic component analysis sub-module and a harmonic parameter calculation sub-module;
acquiring a real-time current signal based on the current signal acquisition sub-module, and acquiring a real-time voltage signal based on the voltage signal acquisition sub-module;
the harmonic component analysis submodule performs waveform measurement and spectrum analysis based on the real-time current signal and the real-time voltage signal to obtain a real-time harmonic component;
the harmonic parameter calculation sub-module determines a real-time harmonic parameter from the real-time harmonic component.
In one embodiment, the harmonic analysis construction unit 5 further comprises:
Carrying out serialization processing on the N groups of deviation harmonic data sets to obtain deviation harmonic extremum;
the first monitoring module further comprises an activation judgment execution sub-module;
constructing an activation judgment parameter based on the deviation harmonic extremum;
the activation judgment execution sub-module synchronizes the activation judgment parameters to;
based on the activation judgment execution submodule, carrying out numerical judgment on the real-time harmonic parameter and the deviation harmonic extremum;
and when the real-time harmonic parameter is larger than the deviation harmonic extremum, sending an activation instruction to activate the secondary filtering module.
In one embodiment, the harmonic analysis construction unit 5 further comprises:
constructing a harmonic control monitoring window;
performing activation frequency statistics of the secondary filtering module based on the harmonic control monitoring window to obtain a plurality of secondary activation frequency information, wherein the plurality of secondary activation frequency information corresponds to a plurality of harmonic control monitoring windows;
performing weight assignment on the secondary activation frequency information according to the time spans of the harmonic control monitoring windows and the current time, and performing weight calculation to obtain steady activation frequency information;
presetting a filter module replacement threshold value, and judging whether the steady-state activation frequency information meets the filter module replacement threshold value;
And if the steady-state activation frequency information meets the filter module replacement threshold value, generating a filter module replacement instruction.
In one embodiment, the secondary control construction unit 7 further comprises:
performing mapping call on the N groups of historical harmonic control parameter information according to the N groups of fitting filtering data sets to obtain N groups of fitting control parameter sets;
and performing supervision training of the standard harmonic controller by adopting the N groups of deviation filtering data sets, the N groups of fitting filtering data sets and the N groups of fitting control parameter sets to obtain the secondary harmonic control analysis sub-network.
Any of the methods or steps described above may be stored as computer instructions or programs in various non-limiting types of computer memories, and identified by various non-limiting types of computer processors, thereby implementing any of the methods or steps described above.
Based on the above-mentioned embodiments of the present invention, any improvements and modifications to the present invention without departing from the principles of the present invention should fall within the scope of the present invention.
Claims (8)
1. A method for harmonic control of a high voltage frequency converter, the method being applied to a harmonic control system of a high voltage frequency converter, the system comprising N sets of harmonic control lines, each set of harmonic control lines comprising a first monitoring module, a second monitoring module and an intelligent filtering module, the method comprising:
Acquiring historical harmonic output information, wherein the historical harmonic output information is acquired through a first monitoring module of an interactive N-group harmonic control circuit, and the historical harmonic output information comprises N groups of historical harmonic data sets;
acquiring historical filtering output information, wherein the historical filtering output information is acquired through a second monitoring module interacting with the N groups of harmonic control lines, and the historical filtering output information comprises N groups of historical filtering data sets;
presetting harmonic constraint parameters, traversing the N groups of historical filtering data sets based on the harmonic constraint parameters, and obtaining N groups of grid filtering data sets and N groups of deviation filtering data sets;
mapping and dividing the N groups of historical harmonic data sets according to the N groups of grid filter data sets and the N groups of deviation filter data sets to obtain N groups of conventional harmonic data sets and N groups of deviation harmonic data sets;
constructing a first-order harmonic control analysis sub-network based on federal learning, and training the first-order harmonic control analysis sub-network based on the N-group conventional harmonic data sets and the N-group filtered data sets;
traversing the N groups of conventional harmonic data sets and the N groups of deviation filtering data sets by taking the N groups of deviation filtering data sets as constraints to obtain N groups of fitting filtering data sets;
Pre-constructing a secondary harmonic control analysis sub-network, and training the secondary harmonic control analysis sub-network based on the N groups of deviation filtering data sets and the N groups of fitting filtering data sets;
and synchronizing the first-level harmonic control analysis sub-network and the second-level harmonic control analysis sub-network to the intelligent filtering module of the N groups of harmonic control lines, and performing harmonic automatic control based on the intelligent filtering module.
2. The method of claim 1, wherein a first order harmonic control analysis sub-network is constructed based on federal learning and training of the first order harmonic control analysis sub-network is performed based on the N-combination lattice filtered data set and the N-set of conventional harmonic data sets, the method further comprising:
the intelligent filtering module comprises a primary filtering module and a secondary filtering module;
acquiring a historical harmonic control parameter set, wherein the historical harmonic control parameter set is obtained by an intelligent filtering module interacting the N groups of harmonic control lines, and the historical harmonic control parameter set comprises N groups of historical harmonic control parameter information;
performing mapping call on the N groups of historical harmonic control parameter information according to the N combination lattice filter data sets to obtain N combination lattice control parameter sets;
Pre-constructing a standard harmonic controller, wherein the standard harmonic controller comprises a decoder and an encoder;
obtaining N harmonic controllers, wherein the N harmonic controllers are obtained by performing supervised training of the standard harmonic controller by adopting the N combined lattice filter data sets, the N groups of conventional harmonic data sets and the N combined lattice control parameter sets;
extracting controller parameters of the N harmonic controllers to obtain N groups of controller parameters;
performing aggregation treatment on the N groups of controller parameters based on a federal aggregation algorithm to obtain optimized controller parameters;
the parameters of the standard harmonic controller are updated by adopting the parameters of the optimized controller, and the primary harmonic control analysis sub-network is obtained;
and N primary filtering modules of the N groups of harmonic control circuits to which the primary harmonic control analysis sub-network is synchronized.
3. The method of claim 2, wherein N harmonic controllers are obtained, wherein the N harmonic controllers are obtained by supervised training of the standard harmonic controller with the N combined lattice filtered data set and the N sets of conventional harmonic data sets, the method further comprising:
Performing mapping call on the N-combination lattice filter data set, the N groups of conventional harmonic data sets and the N-combination lattice control parameter sets to obtain a first qualified filter data set, a first conventional harmonic data set and a first qualified control parameter set;
performing supervised training of a decoder and an encoder of the standard harmonic controller by adopting the first qualified filtering data set, the first conventional harmonic data set and the first qualified control parameter set to obtain a first harmonic controller;
and similarly, performing supervision training of the standard harmonic controller by adopting the N-combination lattice filter data set and the N groups of conventional harmonic data sets to obtain the N harmonic controllers.
4. The method of claim 2, wherein the method further comprises:
the first monitoring module comprises a current signal acquisition sub-module, a voltage signal acquisition sub-module, a harmonic component analysis sub-module and a harmonic parameter calculation sub-module;
acquiring a real-time current signal based on the current signal acquisition sub-module, and acquiring a real-time voltage signal based on the voltage signal acquisition sub-module;
the harmonic component analysis submodule performs waveform measurement and spectrum analysis based on the real-time current signal and the real-time voltage signal to obtain a real-time harmonic component;
The harmonic parameter calculation sub-module determines a real-time harmonic parameter from the real-time harmonic component.
5. The method of claim 4, wherein the method further comprises:
carrying out serialization processing on the N groups of deviation harmonic data sets to obtain deviation harmonic extremum;
the first monitoring module further comprises an activation judgment execution sub-module;
constructing an activation judgment parameter based on the deviation harmonic extremum;
the activation judgment execution sub-module synchronizes the activation judgment parameters to;
based on the activation judgment execution submodule, carrying out numerical judgment on the real-time harmonic parameter and the deviation harmonic extremum;
and when the real-time harmonic parameter is larger than the deviation harmonic extremum, sending an activation instruction to activate the secondary filtering module.
6. The method of claim 5, wherein the method further comprises:
constructing a harmonic control monitoring window;
performing activation frequency statistics of the secondary filtering module based on the harmonic control monitoring window to obtain a plurality of secondary activation frequency information, wherein the plurality of secondary activation frequency information corresponds to a plurality of harmonic control monitoring windows;
Performing weight assignment on the secondary activation frequency information according to the time spans of the harmonic control monitoring windows and the current time, and performing weight calculation to obtain steady activation frequency information;
presetting a filter module replacement threshold value, and judging whether the steady-state activation frequency information meets the filter module replacement threshold value;
and if the steady-state activation frequency information meets the filter module replacement threshold value, generating a filter module replacement instruction.
7. The method of claim 2, wherein a second order harmonic control analysis sub-network is pre-constructed and training of the second order harmonic control analysis sub-network is performed based on the N sets of bias filtered data sets and the N sets of fit filtered data sets, the method further comprising:
performing mapping call on the N groups of historical harmonic control parameter information according to the N groups of fitting filtering data sets to obtain N groups of fitting control parameter sets;
and performing supervision training of the standard harmonic controller by adopting the N groups of deviation filtering data sets, the N groups of fitting filtering data sets and the N groups of fitting control parameter sets to obtain the secondary harmonic control analysis sub-network.
8. A harmonic control system for a high voltage frequency converter, the system comprising:
The system comprises a harmonic information acquisition unit, a first monitoring module and a second monitoring module, wherein the harmonic information acquisition unit is used for acquiring historical harmonic output information, the historical harmonic output information is acquired through the first monitoring module of an interactive N-group harmonic control circuit, and the historical harmonic output information comprises N groups of historical harmonic data sets;
the filtering information acquisition unit is used for acquiring historical filtering output information, wherein the historical filtering output information is acquired through a second monitoring module which interacts with the N groups of harmonic control lines, and the historical filtering output information comprises N groups of historical filtering data sets;
the harmonic constraint setting unit is used for presetting harmonic constraint parameters, traversing the N groups of historical filtering data sets based on the harmonic constraint parameters, and obtaining N groups of grid filtering data sets and N groups of deviation filtering data sets;
the data division execution unit is used for mapping and dividing the N groups of historical harmonic data sets according to the N groups of grid filter data sets and the N groups of deviation filter data sets to obtain N groups of conventional harmonic data sets and N groups of deviation harmonic data sets;
the harmonic analysis construction unit is used for constructing a first-order harmonic control analysis sub-network based on federal learning and training the first-order harmonic control analysis sub-network based on the N-group conventional harmonic data sets and the N-group combination lattice filter data sets;
The data filtering execution unit is used for traversing the N groups of conventional harmonic data sets and the N groups of deviation filtering data sets by taking the N groups of deviation filtering data sets as constraints to obtain N groups of fitting filtering data sets;
the second-level control construction unit is used for pre-constructing a second-level harmonic control analysis sub-network and training the second-level harmonic control analysis sub-network based on the N groups of deviation filtering data sets and the N groups of fitting filtering data sets;
and the harmonic correlation control unit is used for synchronizing the first-level harmonic control analysis sub-network and the second-level harmonic control analysis sub-network to the intelligent filtering modules of the N groups of harmonic control circuits, and carrying out harmonic automatic control based on the intelligent filtering modules.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311362543.3A CN117096956B (en) | 2023-10-20 | 2023-10-20 | Harmonic control method and system of high-voltage frequency converter |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311362543.3A CN117096956B (en) | 2023-10-20 | 2023-10-20 | Harmonic control method and system of high-voltage frequency converter |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117096956A true CN117096956A (en) | 2023-11-21 |
| CN117096956B CN117096956B (en) | 2023-12-26 |
Family
ID=88783364
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311362543.3A Active CN117096956B (en) | 2023-10-20 | 2023-10-20 | Harmonic control method and system of high-voltage frequency converter |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117096956B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102252677A (en) * | 2011-04-18 | 2011-11-23 | 哈尔滨工程大学 | Time series analysis-based variable proportion self-adaptive federal filtering method |
| US20120236606A1 (en) * | 2011-03-18 | 2012-09-20 | George Albert Mazzoli | Method and system for applying power harmonics to secondary loads |
| CN115062668A (en) * | 2022-06-28 | 2022-09-16 | 合肥工业大学 | Harmonic parameter detection method and system based on RAdam optimization width learning |
-
2023
- 2023-10-20 CN CN202311362543.3A patent/CN117096956B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120236606A1 (en) * | 2011-03-18 | 2012-09-20 | George Albert Mazzoli | Method and system for applying power harmonics to secondary loads |
| CN102252677A (en) * | 2011-04-18 | 2011-11-23 | 哈尔滨工程大学 | Time series analysis-based variable proportion self-adaptive federal filtering method |
| CN115062668A (en) * | 2022-06-28 | 2022-09-16 | 合肥工业大学 | Harmonic parameter detection method and system based on RAdam optimization width learning |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117096956B (en) | 2023-12-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114576152B (en) | Water pump state monitoring system, monitoring method and device, electronic equipment and medium | |
| CN104639368A (en) | Method and device for processing faults of communications network equipment | |
| CN103278728A (en) | Method and system for diagnosing interturn short circuit fault of generator rotor | |
| CN113131518B (en) | MLPE photovoltaic system and MLPE equipment detection method thereof | |
| DE102016102320B4 (en) | Dynamically adaptive operation of a device to cope with changes in the quality of electrical power provided | |
| CN117078017A (en) | Intelligent decision analysis system for monitoring power grid equipment | |
| CN109633506B (en) | Data acquisition and checking method and monitoring control system in direct current transmission system | |
| CN106124973A (en) | Converter power transformer load ratio bridging switch on-line condition monitoring method | |
| CN112731827B (en) | Monitoring system for intelligent sensor for power equipment | |
| CN117096956B (en) | Harmonic control method and system of high-voltage frequency converter | |
| CN113285471B (en) | Offshore wind farm subsynchronous oscillation source sensing and positioning method, device and equipment | |
| EP2863551B1 (en) | Method and system for determining an operating status of a network component in a power supply network | |
| CN108693442A (en) | The detection method and system of electric network fault | |
| CN105203915A (en) | Diagnosis system and diagnosis method of loosening defect of power transformer winding | |
| CN116465059A (en) | Energy-saving air conditioner operation supervision system based on data analysis | |
| CN112699104B (en) | Data processing method of intelligent sensor of power equipment and intelligent sensor | |
| CN106019117A (en) | Open-circuit fault diagnosis method of cascaded STATCOM system IGBT | |
| CN1225948C (en) | Method for diagnosing noise failure of accelerator | |
| CN115345519A (en) | Offshore wind power risk assessment method, device, equipment and storage medium | |
| CN109581051A (en) | Adaptive full frequency-domain Wave record method | |
| CN114200213A (en) | Capacitance online monitoring method and device of low-frequency power transmission matrix converter | |
| CN111275938A (en) | Early warning method, device and equipment based on auxiliary equipment and storage medium | |
| CN204615389U (en) | There is the harmonic protection device of record wave energy | |
| Liu et al. | Application of fault diagnosis of high voltage inverter with BP neural network | |
| CN113270942B (en) | Intelligent monitoring device and method for operating state of arc suppression coil |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |