CN116632942A - Series structure micro-grid power coordination control method based on decision tree - Google Patents
Series structure micro-grid power coordination control method based on decision tree Download PDFInfo
- Publication number
- CN116632942A CN116632942A CN202310619732.8A CN202310619732A CN116632942A CN 116632942 A CN116632942 A CN 116632942A CN 202310619732 A CN202310619732 A CN 202310619732A CN 116632942 A CN116632942 A CN 116632942A
- Authority
- CN
- China
- Prior art keywords
- grid
- power
- micro
- phase
- output
- 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 43
- 238000003066 decision tree Methods 0.000 title claims abstract description 37
- 238000010248 power generation Methods 0.000 claims abstract description 72
- 125000003275 alpha amino acid group Chemical group 0.000 claims description 9
- 150000001875 compounds Chemical class 0.000 claims description 9
- 238000012549 training Methods 0.000 claims description 9
- 238000012795 verification Methods 0.000 claims description 7
- 238000004146 energy storage Methods 0.000 claims description 6
- 238000011156 evaluation Methods 0.000 claims description 6
- 238000004088 simulation Methods 0.000 claims description 5
- 101100533306 Mus musculus Setx gene Proteins 0.000 claims description 3
- 239000000969 carrier Substances 0.000 claims description 3
- 238000013507 mapping Methods 0.000 claims description 3
- 238000005530 etching Methods 0.000 claims 1
- 230000002411 adverse Effects 0.000 abstract description 3
- 230000000694 effects Effects 0.000 abstract description 3
- 230000009286 beneficial effect Effects 0.000 abstract 1
- 230000010363 phase shift Effects 0.000 abstract 1
- 230000001105 regulatory effect Effects 0.000 abstract 1
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 230000004069 differentiation Effects 0.000 description 1
Classifications
-
- 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
-
- 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
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- 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
- H02J3/04—Circuit arrangements for ac mains or ac distribution networks for connecting networks of the same frequency but supplied from different sources
- H02J3/06—Controlling transfer of power between connected networks; Controlling sharing of load between connected networks
-
- 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
- H02J3/28—Arrangements for balancing of the load in a network by storage of energy
- H02J3/32—Arrangements for balancing of the load in a network by storage of energy using batteries with converting means
-
- 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
- H02J3/381—Dispersed generators
-
- 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
- H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
- H02J2203/20—Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]
-
- 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
- H02J2300/00—Systems for supplying or distributing electric power characterised by decentralized, dispersed, or local generation
- H02J2300/20—The dispersed energy generation being of renewable origin
- H02J2300/22—The renewable source being solar energy
- H02J2300/24—The renewable source being solar energy of photovoltaic origin
-
- 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
- H02J2300/00—Systems for supplying or distributing electric power characterised by decentralized, dispersed, or local generation
- H02J2300/20—The dispersed energy generation being of renewable origin
- H02J2300/28—The renewable source being wind energy
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Evolutionary Computation (AREA)
- Power Engineering (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention discloses a series structure micro-grid power coordination control method based on a decision tree, which is characterized in that the power of each power generation unit in a grid-connected operation series structure micro-grid is distributed according to the capacity of micro-source output, the adverse effect on grid-connected current caused by the difference of the power generation unit output power is considered, then a characteristic function reflecting the difference of carrier amplitude is constructed, a decision tree model is constructed by taking the characteristic function as main input for judging whether grid-connected current THD is out of limit, if the out of limit condition occurs, the power distribution of the power generation unit is regulated, and grid-connected current control and carrier amplitude phase shift SPWM modulation are combined to realize the coordination control of the power generation unit and the power tracking of a system. The power coordination control method is beneficial to improving the energy utilization rate, and meanwhile, the decision tree model is adopted to pre-judge the grid-connected current quality state caused by power distribution, so that the current quality of the grid-connected operation of the micro-grid with the series structure is ensured.
Description
Technical Field
The invention belongs to the technical field of micro-grid control, and relates to a series structure micro-grid power coordination control method based on a decision tree.
Background
Compared with a parallel structure micro-grid with distributed power supplies and loads transversely distributed on a feeder line, the micro-grid is an effective means for connecting the distributed power supplies to the power grid, and voltage and current output by the micro-grid with the serial structure have the advantages of better sine degree, high output voltage, low dv/dt, high efficiency and the like.
When the micro-grid is in grid-connected operation, a large power grid provides voltage and frequency support, and the main function of the micro-grid is to provide stable power output according to a scheduling instruction of a superior power grid. The output power of the power generation units in the micro-grid with the series structure is related to each other through grid-connected current at the alternating-current side, and compared with the micro-grid with the parallel structure, the power coupling of the power generation units in the micro-grid with the series structure can lead to more complex coordination control. Efficient coordinated control between power generation units is a critical issue if the output of the series-structured microgrid is to be able to track the power command value. Meanwhile, as the micro source in the micro grid with the series structure operates in the maximum power tracking mode, different power is required to be distributed to each power generation unit in order to ensure that 'energy is more spent', the power difference among the power generation units can have adverse effects on grid-connected current, the energy utilization rate of the micro grid with the series structure is reduced, and the quality requirements of power tracking and grid-connected current cannot be met.
Disclosure of Invention
The invention aims to provide a series structure micro-grid power coordination control method based on a decision tree, which can improve the energy utilization rate and the grid-connected current quality of the series structure micro-grid.
In order to achieve the above purpose, the technical scheme adopted by the invention is as follows: a series structure micro-grid power coordination control method based on decision tree comprises the following steps:
1) The series structure micro-grid system runs in a unit power factor grid connection mode, and a power instruction value issued by a superior power grid isP * ;
For single-phase inclusionnMicro-grid with serial structure of micro-sources, if a micro-source or the output power of a micro-sourceP xiS =0, then the output powerP xiS All micro sources of =0 are assigned an output power such that the output powerP xiS The power generation unit where the micro source with the output power of the micro source being=0 participates in power distribution together with the power generation unit where the output power of the micro source is not zero:
is provided withxPhase 1i(i=1, 2, …, n) power generating units G Uxi The output power of the micro source isP xiS Power generation unit G after power distribution Uxi Output power of (2)P xiG The method comprises the following steps:
after power distributionxThe total power output by all the power generation units in the phase meets the following conditions:
after distributing power to each power generation unit according to (1), the modulation ratio of H-bridge MSI connected to each power generation unitM xi The method comprises the following steps:
in the method, in the process of the invention,M e representing equivalent modulation ratios corresponding to all power generation units when the power command value is equally divided;nthe number of micro sources in the micro grid with the single-phase series structure;
three-phase series structure micro-grid systemThe system adopts carrier amplitude-varying phase-shifting SPWM modulation, and obtains three-phase modulation wave signals under grid-connected current controlU xm On the basis of the method, the device comprises the steps of,M xi by varying the respective carrier amplitudeU xic The realization of the method is realized in that,U xic =U m /M xi (4)
in the formula (4), the amino acid sequence of the compound,U m for modulating the amplitude of the wave signal;
carrier amplitudeU xic Is a constraint on (c):
in the formula (5), the amino acid sequence of the compound,U c the carrier amplitude of MSI when the power command value is equally divided;
2) The difference between the output powers of the power generation units is represented by the input characteristics of the difference between the output powers of the power generation units, wherein the carrier amplitudes of the MSIs corresponding to the power generation units are differentβ x Is thatxThe variance of the reciprocal of the phase carrier amplitude, namely:
3) Generating a grid-connected current THD threshold-crossing decision tree model by adopting a classification regression tree, wherein the model is input asX={P * ,β x Output is recorded asY=[T x ],T x The method comprises the steps of (1) a grid-connected current THD out-of-limit state of each phase of the series structure micro grid;
the multi-label output of the decision tree model is regarded as a three-bit binary number, and the three-bit binary number is converted into a corresponding decimal number to be used as multi-classification outputY' converting the multi-label classification task into a multi-classification task; generating a large amount of triangular carrier amplitude data meeting the formula (5) through a python random function, and simultaneously calculating the variance of the inverse of the carrier amplitude, namely the input characteristic, through the formula (6)β x Selecting a feature to be enteredβ x Data distributed uniformly for running at different power command valuesP * And under the constraint of formula (5)β x Respectively simulating the series structure micro-grids, setting the threshold value of the grid-connected current THD, obtaining each sample output category according to the grid-connected current THD condition, simulating to obtain sample data, taking part of the sample data as training sample data, and taking the rest of the sample data as verification sample data; training sample data through the decision tree model, establishing a mapping relation between input characteristics and grid-connected current THD out-of-limit states, and evaluating the performance of the decision tree model on verification sample data through two evaluation indexes of macro average and classification accuracy until the performance of the decision tree model on the sample data meets the requirements of the evaluation indexes, so as to obtain a trained decision tree model;
when in online use, the characteristic data setX={P * ,β x Inputting the model into a trained decision tree, predicting the grid-connected current THD out-of-limit state caused by corresponding power distribution, and obtaining model outputY′;
If it isY' =0, i.e. there is no grid-connected current THD out of limit, control signalU OL Put 1, the first switch S 1 Switching off the carrier amplitude at the previous momentU cxi,t-1 And turning on the carrier amplitude obtained by the time coordination controlU cxi,t, ;
If it isY' not equal to 0, i.e. when grid-connected current THD out-of-limit exists, the control signalU OL Setting 0, the carrier amplitude is not changed for a while and still usedU cxi,t-1 At the same time, a second switch S PA Closing, and starting power adjustment of the power generation unit;
the specific process of power adjustment of the power generation unit comprises the following steps: carrier amplitude comparison table T, t= { is established by simulationP * :U isc },U isc Representing carrier amplitude, the carrier amplitudeU isc Can ensure the currentP * THD of the lower grid-connected current is maximum but not out of limit; the current carrier amplitudeU cxi Carrier amplitude in a look-up tableU sci Respectively sorting the carriers from big to small, and sequentially sorting the carrier amplitudes according to the sorted orderU isc Assigned to the carrier amplitudeU isc Current carrier amplitude with same rank numberU xic Amplitude of three-phase carrier waveU xic Adjust to the corresponding table in the comparison tableP * Corresponding size, adjusted power generation unit G Uxi The output power variation is:
in the formula (7), the amino acid sequence of the compound,U dc MSI direct current side voltage is in unit V;I x is thatxPhase-parallel network current effective value;φ x is thatxA phase difference between the phase-parallel network current and the fundamental component of the output voltage;
the output power variation delta P Gxi The power adjustment of the power generation units is realized by controlling the charge and discharge of the energy storage devices;
control signal after power adjustmentP are And control signalU OL Are all set to 1, and the carrier amplitude U after adjustment sci Carrier amplitude U obtained as the time coordination control cxi,t ;
4) Obtaining three-phase modulated wave signals by direct current controlu mx The method comprises the steps of carrying out a first treatment on the surface of the And sending the modulated wave signals of each phase and the amplitude of the triangular carrier to a carrier amplitude-varying phase-shifting SPWM (sinusoidal pulse width modulation) module, and generating pulse signals to drive a three-phase switching tube to work so as to realize power coordination of a power generation unit and power tracking of a system.
The coordination control method is based on the characteristics of the micro-grid with the series structure, the 'energy' thought is taken as guidance to formulate a power distribution method, meanwhile, the influence of power generation unit output power differentiation on grid-connected current is fully considered, a THD (total harmonic distortion) out-of-limit decision tree of grid-connected current is constructed to realize real-time prediction of current quality state, and power distribution of the power generation unit under the condition of THD out-of-limit of the grid-connected current is adjusted to realize coordination control of power of the power generation unit and power tracking of a system, so that the micro-grid with the series structure runs in a grid-connected mode with higher electric energy quality.
The coordination control method solves the problem that the power difference between the power generation units has adverse effect on grid-connected current, improves the energy utilization rate, simultaneously meets the requirements of power tracking and the quality of the grid-connected current of the micro-grid of the series structure, reduces the pollution to the grid, and is suitable for the coordination control of the power generation units during the grid-connected operation of the micro-grid of the series structure.
Drawings
Fig. 1 is a topological diagram of a series structure micro-grid in a grid-connected operation mode.
Fig. 2 is a schematic diagram of the control method of the present invention.
Detailed Description
The invention will now be described in detail with reference to the drawings and examples.
Examples
The three-phase series structure micro-grid system topology structure in the grid-connected operation mode is shown in fig. 1. The topological structure comprises an a-phase series structure micro-grid, a b-phase series structure micro-grid and a c-phase series structure micro-grid;
the micro-grid with the a-phase series structure, the micro-grid with the b-phase series structure and the micro-grid with the c-phase series structure comprise a plurality of photovoltaic micro-sources and a plurality of wind power micro-sources, and all the photovoltaic micro-sources and all the wind power micro-sources are respectively provided with an energy storage device and a converter. The photovoltaic micro-source, the energy storage device and the converter which are arranged on the photovoltaic micro-source form a first power generation unit, and the wind micro-source, the energy storage device and the converter which are arranged on the wind micro-source form a second power generation unit. Each power generation unit is connected to an H-bridge Micro Source Inverter (MSI). All H-bridge MSI output ends in the micro-grid with the in-phase series structure are connected in series to form a multi-level system single-phase output voltage in a carrier phase-shifting SPWM modulation mode,
the number of the power generation units in the micro-grid with the a-phase series structure, the number of the power generation units in the micro-grid with the b-phase series structure and the number of the power generation units in the micro-grid with the c-phase series structure are the same.
The other output end of the first H-bridge micro-source inverter a in the micro-grid with the a-phase series structure is connected with the other output end of the first H-bridge micro-source inverter c in the micro-grid with the c-phase series structure; the other output end of the first H-bridge micro-source inverter b in the micro-grid with the b-phase series structure is connected with a filter; the other output end of the last H-bridge micro-source inverter a in the micro-grid with the a-phase series structure is respectively connected with the other output end of the last H-bridge micro-source inverter b in the micro-grid with the b-phase series structure and the other output end of the last H-bridge micro-source inverter c in the micro-grid with the c-phase series structure.
The multi-level voltage output by the three-phase series structure micro-grid system is filtered by a filter to obtain electric energy with higher sine degree, and the electric energy is connected to a power grid at a public coupling point (point of common coupling, PCC). And the voltage and current at the PCC are collected, and the control of the three-phase H-bridge micro-source inverter is realized after passing through the controller and the modulation circuit, so that the system output meets the power grid dispatching and load requirements.
The coordination control method is used for carrying out coordination control on the three-phase series structure micro-grid system shown in fig. 1, the control principle is shown in fig. 2, and the coordination control method specifically comprises the following steps:
the three-phase series structure micro-grid system runs in a unit power factor grid connection mode, and a power instruction value issued by a superior power grid is as followsP * Power allocation: distributing power to a power generation unit where the photovoltaic micro source is positioned or a power generation unit where the wind micro source is positioned according to the ratio of the output power of the photovoltaic micro source or the output power of the wind micro source in the total power of all micro source power, and simultaneously ensuring that the total power output of a three-phase series structure micro grid system tracks a power instruction valueP * 。
1) The series structure micro-grid system runs in a unit power factor grid connection mode, and a power instruction value issued by a superior power grid isP * The method comprises the steps of carrying out a first treatment on the surface of the For single-phase inclusionnMicro-grid with serial structure of micro-sources, if a micro-source or the output power of a micro-sourceP xiS =0, e.g. photovoltaic micro-source operating at night, to avoid its exit from operation, output powerP xiS All micro sources=0 assign an output power of 0.1kw to 0.5kw, to make output powerP xiS The power generation unit where the micro source with the output power of the micro source being=0 participates in power distribution together with the power generation unit where the output power of the micro source is not zero:
is provided withxPhase (in this example)xA, b or c)i(i=1, 2, …, n) power generating units G Uxi The output power of the micro source isP xiS Power generation unit G after power distribution Uxi Output power of (2)P xiG The method comprises the following steps:
after power distributionxThe total power output by all the power generation units in the phase meets the following conditions:
as seen from the formula (2), after the power is distributed according to the formula (1), the total output power of the power generation units in the single-phase series-structure micro-grid is equal to the power command value, so that the three-phase series-structure micro-grid system can track the power command value.
After distributing power to each power generation unit according to (1), the modulation ratio of H-bridge MSI connected to each power generation unitM xi The method comprises the following steps:
in the method, in the process of the invention,M e representing equivalent modulation ratios corresponding to all power generation units when the power command value is equally divided;nthe number of micro sources in the micro grid with the single-phase series structure;
the three-phase series structure micro-grid system adopts carrier amplitude-varying phase-shifting SPWM modulation and obtains three-phase modulation wave signals under direct current controlU xm (xBased on a, b or c),M xi by varying the respective carrier amplitudeU xic The realization is that:
U xic =U m /M xi (4)
in the formula (4), the amino acid sequence of the compound,U m for modulating the amplitude of the wave signal.
Carrier amplitudeU xic The following formula serves as a constraint:
in the formula (5), the amino acid sequence of the compound,U c the carrier amplitude of MSI when the power command value is equally divided;
when the carrier amplitude meets the constraint condition of the formula (5), the power generation unit in the single-phase series structure micro-grid outputs the distributed power according to the output capacity of the contained micro-source, the output power of the single-phase series structure micro-grid is equal to the instruction power, and the three-phase system outputs balance.
2) Constructing input features for reflecting differences between output powers of power generation unitsβ x (x=a, b or c):
the difference between the output power of each power generation unit is represented by the different carrier amplitudes of the MSIs corresponding to the power generation units, and the input characteristics areβ x Is thatxThe variance of the reciprocal of the phase carrier amplitude, namely:
3) Constructing a total harmonic distortion (Total Harmonic Distortion, THD) out-of-limit decision tree of grid-connected current:
generating a grid-connected current THD threshold-crossing decision tree model by adopting a classification regression tree, wherein the model is input asX={P * ,β a ,β b ,β c Output is recorded asY=[T ca ,T cb ,T cc ],T ca 、T cb AndT cc grid-connected current THD out-of-limit states of a phase, b phase and c phase respectively. When the phase a, phase b or phase c parallel network current THD is not limited, the phase a, phase b or phase c parallel network current THD corresponds toT ca 、T cb Or (b)T cc Taking 0; when the phase a, the phase b or the phase c parallel network current THD is more limited, the phase a, the phase b or the phase c parallel network current THD corresponds toT ca 、T cb Or (b)T cc Taking 1. Due to the decision tree modelEach tag of the output is a classifier, so that the multi-tag output can be regarded as a three-bit binary number, and the three-bit binary number is converted into a corresponding decimal number to be used as the multi-classification outputY' the original multi-label classification task can be converted into a multi-classification task, so that a model is simplified;
generating a large amount of triangular carrier amplitude data meeting the formula (5) through a python random function, and simultaneously calculating the variance of the inverse of the carrier amplitude, namely the input characteristic, through the formula (6)β x Selecting a feature to be enteredβ x Data distributed uniformly for running at different power command valuesP * And under the constraint of formula (5)β x Micro-grid with serial structuren10) respectively carrying out simulation, setting the grid-connected current THD to exceed 5% as out-of-limit, obtaining each sample output category according to the grid-connected current THD condition, carrying out simulation to obtain 400 groups of sample data, taking 70% of the sample data as a training sample set (280 groups) and 30% of the sample data as a verification sample set (120 groups), training the training sample data through a decision tree model, establishing a mapping relation between input characteristics and the grid-connected current THD out-of-limit state, and evaluating the performance of the decision tree model on the verification sample set through two evaluation indexes of macro average and classification accuracy until the performance of the decision tree model on the sample set meets the requirements of the evaluation indexes, and completing offline model training to obtain a trained decision tree model;
when in online use, the characteristic data setX={P * ,β a ,β b ,β c Inputting the model into a trained decision tree, predicting the grid-connected current THD out-of-limit state caused by corresponding power distribution, and obtaining model outputY′;
If it isY' =0, i.e. there is no grid-connected current THD out of limit, the control signalU OL Put 1, the first switch S 1 Switching off the carrier amplitude at the previous momentU cxi,t-1 And turning on the carrier amplitude obtained by the time coordination controlU cxi,t ;
If it isY' not equal to 0, i.e. if THD is out of limit, thenControl signalU OL Setting 0, the carrier amplitude is not changed for a while and still usedU cxi,t-1 At the same time, a second switch S PA Closing, and starting power adjustment of the power generation unit;
the specific process of power adjustment of the power generation unit comprises the following steps: carrier amplitude comparison table T, t= { is established by simulationP * :U isc },U isc Representing carrier amplitude, the carrier amplitudeU isc Can ensure the currentP * THD of the lower grid-connected current is maximum but not out of limit; the current carrier amplitudeU cxi Carrier amplitude in a look-up tableU sci Respectively sorting the carriers from big to small, and sequentially sorting the carrier amplitudes according to the sorted orderU isc Assigned to the carrier amplitudeU isc Current carrier amplitude with same rank numberU xic Amplitude of three-phase carrier waveU xic Adjust to the corresponding table in the comparison tableP * Corresponding size, adjusted power generation unit G Uxi The output power variation is:
in the formula (7), the amino acid sequence of the compound,U dc MSI direct current side voltage is in unit V;I x is thatxPhase-parallel network current effective value;φ x is thatxA phase difference between the phase-parallel network current and the fundamental component of the output voltage;
the output power variation delta P Gxi The power adjustment of the power generation units is realized by controlling the charge and discharge of the energy storage devices;
control signal after power adjustmentP are And control signalU OL Are all set to 1, and the carrier amplitude U after adjustment sci Carrier amplitude U obtained as the time coordination control cxi,t ;
4) Obtaining three-phase modulated wave signals by direct current controlu mx (x=a, b or c), where, andthe network current regulator adopts a quasi-proportional resonance controller; meanwhile, the triangular carrier amplitude of each micro-source inverter is obtained through the step 3), each phase of modulated wave signal and the triangular carrier amplitude are sent into a carrier amplitude-varying phase-shifting SPWM (sinusoidal pulse width modulation) module, and a pulse signal is generated to drive a three-phase switching tube to work, so that power coordination of a power generation unit and system power tracking are realized.
The invention can ensure the power tracking of the system while realizing the coordination control of the power generation unit, the realization of the system level power tracking is composed of two parts of direct current control and power coordination control of the power generation unit, and the power coordination control of the power generation unit comprises the steps of power distribution of the power generation unit, THD out-of-limit pre-judgment of grid current, power adjustment under the THD out-of-limit condition and the like.
Claims (6)
1. A series structure micro-grid power coordination control method based on a decision tree is characterized by comprising the following steps:
1) The series structure micro-grid system runs in a unit power factor grid connection mode, and a power instruction value issued by a superior power grid isP * ;
For single-phase inclusionnMicro-grid with serial structure of micro-sources, if a micro-source or the output power of a micro-sourceP xiS =0, then the output powerP xiS All micro sources of =0 are assigned an output power such that the output powerP xiS The power generation unit where the micro source with the output power of the micro source being=0 participates in power distribution together with the power generation unit where the output power of the micro source is not zero:
is provided withxPhase 1i(i=1, 2, …, n) power generating units G Uxi The output power of the micro source isP xiS Power generation unit G after power distribution Uxi Output power of (2)P xiG The method comprises the following steps:
after power distributionxTotal work output by all power generating units in phaseThe rate satisfies:
after distributing power to each power generation unit according to (1), the modulation ratio of H-bridge MSI connected to each power generation unitM xi The method comprises the following steps:
in the method, in the process of the invention,M e representing equivalent modulation ratios corresponding to all power generation units when the power command value is equally divided;nthe number of micro sources in the micro grid with the single-phase series structure;
the three-phase series structure micro-grid system adopts carrier amplitude-varying phase-shifting SPWM modulation, and obtains three-phase modulated wave signals under grid-connected current controlU xm On the basis of the method, the device comprises the steps of,M xi by varying the respective carrier amplitudeU xic The realization of the method is realized in that,U xic =U m /M xi (4)
in the formula (4), the amino acid sequence of the compound,U m for modulating the amplitude of the wave signal;
carrier amplitudeU xic Is a constraint on (c):
in the formula (5), the amino acid sequence of the compound,U c the carrier amplitude of MSI when the power command value is equally divided;
2) The difference between the output powers of the power generation units is represented by the input characteristics of the difference between the output powers of the power generation units, wherein the carrier amplitudes of the MSIs corresponding to the power generation units are differentβ x Is thatxThe variance of the reciprocal of the phase carrier amplitude, namely:
3) Generating a grid-connected current THD threshold-crossing decision tree model by adopting a classification regression tree, wherein the model is input asX={P * , β x Output is recorded asY=[T x ],T x The method comprises the steps of (1) a grid-connected current THD out-of-limit state of each phase of the series structure micro grid;
the multi-label output of the decision tree model is regarded as a three-bit binary number, and the three-bit binary number is converted into a corresponding decimal number to be used as multi-classification outputY' converting the multi-label classification task into a multi-classification task; generating a large amount of triangular carrier amplitude data meeting the formula (5) through a python random function, and simultaneously calculating the variance of the inverse of the carrier amplitude, namely the input characteristic, through the formula (6)β x Selecting a feature to be enteredβ x Data distributed uniformly for running at different power command valuesP * And under the constraint of formula (5)β x Respectively simulating the series structure micro-grids, setting the threshold value of the grid-connected current THD, obtaining each sample output category according to the grid-connected current THD condition, simulating to obtain sample data, taking part of the sample data as training sample data, and taking the rest of the sample data as verification sample data; training sample data through the decision tree model, establishing a mapping relation between input characteristics and grid-connected current THD out-of-limit states, and evaluating the performance of the decision tree model on verification sample data through two evaluation indexes of macro average and classification accuracy until the performance of the decision tree model on the sample data meets the requirements of the evaluation indexes, so as to obtain a trained decision tree model;
when in online use, the characteristic data setX={P * , β x Inputting the model into a trained decision tree, predicting the grid-connected current THD out-of-limit state caused by corresponding power distribution, and obtaining model outputY′;
If it isY' =0, i.e. there is no grid-connected current THD out of limit, control signalU OL Put 1, the first switch S 1 Switching off the carrier amplitude at the previous momentU cxi,t-1 While switching onCarrier amplitude obtained by etching coordination controlU cxi,t, ;
If it isY' not equal to 0, i.e. when grid-connected current THD out-of-limit exists, the control signalU OL Setting 0, the carrier amplitude is not changed for a while and still usedU cxi,t-1 At the same time, a second switch S PA Closing, and starting power adjustment of the power generation unit;
the specific process of power adjustment of the power generation unit comprises the following steps: carrier amplitude comparison table T, t= { is established by simulationP * : U isc },U isc Representing carrier amplitude, the carrier amplitudeU isc Can ensure the currentP * THD of the lower grid-connected current is maximum but not out of limit; the current carrier amplitudeU cxi Carrier amplitude in a look-up tableU sci Respectively sorting the carriers from big to small, and sequentially sorting the carrier amplitudes according to the sorted orderU isc Assigned to the carrier amplitudeU isc Current carrier amplitude with same rank numberU xic Amplitude of three-phase carrier waveU xic Adjust to the corresponding table in the comparison tableP * Corresponding size, adjusted power generation unit G Uxi The output power variation is:
in the formula (7), the amino acid sequence of the compound,U dc MSI direct current side voltage is in unit V;I x is thatxPhase-parallel network current effective value;φ x is thatxA phase difference between the phase-parallel network current and the fundamental component of the output voltage;
the output power variation delta P Gxi The power adjustment of the power generation units is realized by controlling the charge and discharge of the energy storage devices;
control signal after power adjustmentP are And control signalU OL Are all set to 1, and the carrier amplitude U after adjustment sci As the time coordinationCarrier amplitude U obtained by modulation control cxi,t ;
4) Obtaining three-phase modulated wave signals by direct current controlu mx The method comprises the steps of carrying out a first treatment on the surface of the And sending the modulated wave signals of each phase and the amplitude of the triangular carrier to a carrier amplitude-varying phase-shifting SPWM (sinusoidal pulse width modulation) module, and generating pulse signals to drive a three-phase switching tube to work so as to realize power coordination of a power generation unit and power tracking of a system.
2. The method for coordinated control of power in a series micro-grid based on decision tree as recited in claim 1, wherein in said step 1), if the output power of a micro-source or micro-sources isP xiS =0, then the output powerP xiS All micro sources=0 assign an output power of 0.1kw to 0.5 kw.
3. The method for coordinated control of power of a micro-grid with a serial structure based on decision tree as claimed in claim 1, wherein in the step 1), when the carrier amplitude satisfies the constraint condition of the formula (5), the power generating unit in the micro-grid with a single-phase serial structure outputs the distributed power according to the output capacity of the micro-source contained, and the output power of the micro-grid with a single-phase serial structure is equal to the command power, and the three-phase system outputs a balance.
4. The method for controlling power coordination of a series-structured micro-grid based on decision tree as set forth in claim 1, wherein in the step 3), when the grid-connected current THD of a micro-grid of a certain phase of series-structured is not out of limit, the grid-connected current THD out of limit state corresponding to the micro-grid of the certain phase of series-structured is taken to be 0; when the grid-connected current THD of the micro-grid of the certain phase series structure is out of limit, taking 1 corresponding to the grid-connected current THD out-of-limit state of the micro-grid of the certain phase series structure.
5. The method for coordinated control of micro-grid power based on a series structure of decision tree according to claim 1, wherein in the step 3), the threshold value of the grid-connected current THD is set to be 5%, and the threshold value exceeding 5% is the threshold value.
6. The method for coordinated control of power of a series-structured micro-grid based on decision tree according to claim 1, wherein in the step 3), 70% of the sample data is used as a training sample set and 30% is used as a verification sample set.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310619732.8A CN116632942B (en) | 2023-05-30 | 2023-05-30 | Series structure micro-grid power coordination control method based on decision tree |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310619732.8A CN116632942B (en) | 2023-05-30 | 2023-05-30 | Series structure micro-grid power coordination control method based on decision tree |
Publications (2)
Publication Number | Publication Date |
---|---|
CN116632942A true CN116632942A (en) | 2023-08-22 |
CN116632942B CN116632942B (en) | 2023-10-10 |
Family
ID=87609531
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310619732.8A Active CN116632942B (en) | 2023-05-30 | 2023-05-30 | Series structure micro-grid power coordination control method based on decision tree |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116632942B (en) |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103296700A (en) * | 2013-04-15 | 2013-09-11 | 浙江大学 | Interconnection-line-less compensation control method of microgrid harmonic waves and idle currents |
CN107154646A (en) * | 2017-06-08 | 2017-09-12 | 兰州理工大学 | Tandem type microgrid power coordination approach based on micro- source peak power output |
CN111864734A (en) * | 2020-07-21 | 2020-10-30 | 兰州理工大学 | Method for evaluating interaction stability of series-structure microgrid and power grid |
-
2023
- 2023-05-30 CN CN202310619732.8A patent/CN116632942B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103296700A (en) * | 2013-04-15 | 2013-09-11 | 浙江大学 | Interconnection-line-less compensation control method of microgrid harmonic waves and idle currents |
CN107154646A (en) * | 2017-06-08 | 2017-09-12 | 兰州理工大学 | Tandem type microgrid power coordination approach based on micro- source peak power output |
CN111864734A (en) * | 2020-07-21 | 2020-10-30 | 兰州理工大学 | Method for evaluating interaction stability of series-structure microgrid and power grid |
Non-Patent Citations (3)
Title |
---|
DING YINGJIE 等: ""A Study of Over-modulation Impact on the MSI Series Micro-Grid Output Voltage"", 《2019 4TH ASIA CONFERENCE ON POWER AND ELECTRICAL ENGINEERING (ACPEE 2019)》, pages 1 - 9 * |
王兴贵;李世洁;王海亮;: "一种串联型微电网功率协调控制策略", 自动化与仪器仪表, no. 07, pages 210 - 212 * |
王兴贵;杨维满;: "一种微源逆变器串联连接型微网特性研究", 电力系统保护与控制, no. 21, pages 134 - 140 * |
Also Published As
Publication number | Publication date |
---|---|
CN116632942B (en) | 2023-10-10 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Ramaraju et al. | Design and experimental investigation on VL-MLI intended for half height (HH) method to improve power quality using modified particle swarm optimization (MPSO) algorithm | |
CN110149065B (en) | Buck-boost switched capacitor multi-level inverter and modulation method thereof | |
CN110048629B (en) | Single-input switched capacitor multi-level inverter and modulation method thereof | |
Rao et al. | Grid Connected Distributed Generation System with High Voltage Gain Cascaded DC-DC Converter Fed Asymmetric Multilevel Inverter Topology. | |
Busarello et al. | Understanding the staircase modulation strategy and its application in both isolated and grid-connected asymmetric cascaded H-bridge multilevel inverters | |
Kitmo et al. | Optimization of the photovoltaic systems on the North Cameroon interconnected electrical grid | |
Anjaneya Vara Prasad et al. | An Overview on Multi‐Level Inverter Topologies for Grid‐Tied PV System | |
CN114530883A (en) | Power control method, device and system of light storage integrated grid-connected inverter | |
CN103904927B (en) | Additional direct-current voltage control method for modular multi-level current converter | |
Bharathi et al. | Power flow control based on bidirectional converter for hybrid power generation system using microcontroller | |
CN116632942B (en) | Series structure micro-grid power coordination control method based on decision tree | |
Jitta et al. | Switching angle design for pulse width modulation AC voltage controller using genetic algorithm and distributed artificial neural network | |
CN111509830B (en) | Topological structure of miniature photovoltaic/energy storage intelligent power station | |
Gopinath et al. | A pathway to explore the hidden specialty in the design of fifteen level inverter in grid connected PV system | |
Gurgi et al. | Harmonic reduction for grid-connected photovoltaic system based on multilevel inverter | |
Silva et al. | A single-phase five-level grid-connected inverter for photovoltaic applications | |
Sreekumar et al. | Optimisation of Harmonic Droop Coefficient to Minimize Power Losses in an Islanded Microgrid | |
Rekioua et al. | Power control-based fuzzy and modulated hysteresis methods for micro-grid using a photovoltaic system | |
CN113972850B (en) | NL-SPWM-based MMC double-bridge arm complementary hybrid modulation method | |
Ralhan et al. | Comparative Analysis of FOPI Controller using Eagle Perching Optimization for Fuel Cell connected Distributed Generation System | |
Gakhar et al. | Power Quality Improvement in a Grid-Tied SPV System Using Fractional-Order Proportional Integral Controller | |
CN113328427B (en) | Direct current power supply system and power supply method based on deep learning | |
Muthukaruppasamy et al. | A Symmetric Multi-Level Cascaded H-Bridge Inverter for Renewable Energy Integration | |
CN220874229U (en) | Photovoltaic energy storage device and system | |
Jing et al. | Comparative Analysis of the SHEPWM and SHMPWM Techniques for the Three-Level NPC Inverter based on the THD of Voltage and Current |
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 |