CN114123279A - Energy storage state of charge reference value prediction method and system of wind power energy storage hybrid system - Google Patents
Energy storage state of charge reference value prediction method and system of wind power energy storage hybrid system Download PDFInfo
- Publication number
- CN114123279A CN114123279A CN202111415712.6A CN202111415712A CN114123279A CN 114123279 A CN114123279 A CN 114123279A CN 202111415712 A CN202111415712 A CN 202111415712A CN 114123279 A CN114123279 A CN 114123279A
- Authority
- CN
- China
- Prior art keywords
- energy storage
- power
- charge
- reference value
- state
- 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
- 238000004146 energy storage Methods 0.000 title claims abstract description 231
- 238000000034 method Methods 0.000 title claims abstract description 35
- 238000012549 training Methods 0.000 claims abstract description 47
- 238000009499 grossing Methods 0.000 claims abstract description 42
- 238000010801 machine learning Methods 0.000 claims abstract description 16
- 238000013507 mapping Methods 0.000 claims abstract description 5
- 230000006870 function Effects 0.000 claims description 26
- 210000002569 neuron Anatomy 0.000 claims description 13
- 238000004590 computer program Methods 0.000 claims description 12
- 238000012795 verification Methods 0.000 claims description 9
- 230000005611 electricity Effects 0.000 claims description 7
- 238000012545 processing Methods 0.000 claims description 7
- 238000003860 storage Methods 0.000 claims description 7
- 230000005284 excitation Effects 0.000 claims description 6
- 238000012886 linear function Methods 0.000 claims description 6
- 238000005457 optimization Methods 0.000 claims description 5
- 238000010606 normalization Methods 0.000 claims description 3
- 238000012546 transfer Methods 0.000 claims description 3
- 230000010354 integration Effects 0.000 claims 1
- 230000000694 effects Effects 0.000 abstract description 11
- 238000010586 diagram Methods 0.000 description 19
- 238000013528 artificial neural network Methods 0.000 description 17
- 238000011217 control strategy Methods 0.000 description 13
- 238000007599 discharging Methods 0.000 description 11
- 238000004088 simulation Methods 0.000 description 11
- 238000010248 power generation Methods 0.000 description 5
- 238000001914 filtration Methods 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 230000009467 reduction Effects 0.000 description 3
- 230000006583 body weight regulation Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 230000005484 gravity Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000035515 penetration Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
Images
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
- H02J3/28—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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/048—Activation functions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- 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
- H02J3/004—Generation forecast, e.g. methods or systems for forecasting future energy generation
-
- 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
- H02J3/24—Arrangements for preventing or reducing oscillations of power in 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
- H02J3/38—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
- H02J7/00—Circuit arrangements for charging or depolarising batteries or for supplying loads from batteries
- H02J7/0029—Circuit arrangements for charging or depolarising batteries or for supplying loads from batteries with safety or protection devices or circuits
- H02J7/00302—Overcharge protection
-
- 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
- H02J7/00—Circuit arrangements for charging or depolarising batteries or for supplying loads from batteries
- H02J7/0029—Circuit arrangements for charging or depolarising batteries or for supplying loads from batteries with safety or protection devices or circuits
- H02J7/00306—Overdischarge protection
-
- 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
- H02J7/00—Circuit arrangements for charging or depolarising batteries or for supplying loads from batteries
- H02J7/007—Regulation of charging or discharging current or voltage
- H02J7/00712—Regulation of charging or discharging current or voltage the cycle being controlled or terminated in response to electric parameters
-
- 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
- H02J2300/28—The renewable source being wind energy
-
- 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
- Y02E10/00—Energy generation through renewable energy sources
- Y02E10/70—Wind energy
- Y02E10/76—Power conversion electric or electronic aspects
Landscapes
- Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- General Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Tourism & Hospitality (AREA)
- General Engineering & Computer Science (AREA)
- Molecular Biology (AREA)
- Mathematical Physics (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- General Business, Economics & Management (AREA)
- Computing Systems (AREA)
- Artificial Intelligence (AREA)
- Marketing (AREA)
- Life Sciences & Earth Sciences (AREA)
- Primary Health Care (AREA)
- Water Supply & Treatment (AREA)
- Public Health (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention discloses a method and a system for predicting an energy storage state of charge reference value of a wind power energy storage hybrid system, wherein the method comprises the steps of obtaining an average power predicted value of the wind power energy storage hybrid system in two continuous time periods before and after the latest time period; and inputting the average power predicted values of the two continuous time periods into a machine learning network model with two inputs and one output, which completes training in advance, so as to obtain the energy storage state of charge reference value of the energy storage device in the wind power energy storage hybrid system, wherein the machine learning network model with two inputs and one output is trained in advance to establish the mapping relation between the average power predicted values of the two continuous time periods and the energy storage state of charge reference value of the energy storage device. The invention can realize the optimal smoothing effect on the power fluctuation of the wind power plant on the premise of certain power and capacity configuration of the battery energy storage system, and simultaneously can also prevent the battery energy storage system from generating the condition of overcharge or deep discharge.
Description
Technical Field
The invention relates to the technical field of power systems, in particular to a method and a system for predicting an energy storage state of charge reference value of a wind power energy storage hybrid system.
Background
The wind power energy storage hybrid system is a wind power generation system with an energy storage device. The wind power generation system cannot continuously and stably output electric energy due to changes of natural conditions such as wind speed and wind direction, and under the condition of high wind power penetration power, the wind power generation system can greatly affect the electric energy quality and stability of a power grid. Therefore, the energy storage device with certain capacity is configured in the system to play the roles of smoothing power fluctuation, maintaining power generation/load dynamic balance and keeping voltage/frequency stable, thereby realizing the safe, economic, efficient and high-quality operation of the wind power generation system. In terms of the technical performance of the energy storage system, the larger the capacity configuration is, the better the smoothing effect on the power fluctuation of the wind power system is, but the investment cost of the system is increased at the same time, and the economic requirement cannot be well met. Therefore, how to improve the technical performance of the energy storage system in the wind power system through the optimization control of the energy storage system with certain capacity becomes a problem to be solved urgently at present.
Disclosure of Invention
The technical problems to be solved by the invention are as follows: aiming at the problems in the prior art, the invention provides a method and a system for predicting the energy storage state-of-charge reference value of a wind power energy storage hybrid system.
In order to solve the technical problems, the invention adopts the technical scheme that:
a method for predicting an energy storage state of charge reference value of a wind power energy storage hybrid system comprises the following steps:
1) acquiring the average power predicted value of the wind power energy storage hybrid system in the latest two continuous time periods;
2) and inputting the average power predicted values of the two continuous time periods into a machine learning network model with two inputs and one output, which completes training in advance, so as to obtain the energy storage state of charge reference value of the energy storage device in the wind power energy storage hybrid system, wherein the machine learning network model with two inputs and one output is trained in advance to establish the mapping relation between the average power predicted values of the two continuous time periods and the energy storage state of charge reference value of the energy storage device.
Optionally, the two-input-single-output machine learning network model is a two-input-single-output single hidden layer BP network.
Optionally, the single hidden layer BP network includes an input layer, a hidden layer and an output layer, where the input layer includes two neurons for inputting the predicted average power values of two consecutive periods before and after the input layer, and the hidden layer includes 6 neurons for outputting X1～X6And the output layer is used for outputting the obtained energy storage charge state reference value of the energy storage device.
Optionally, the input layer and the hidden layer are connected by a hyperbolic tangent excitation function tansig, and a function expression of the hidden layer is as follows:
in the above formula, tanh represents a hyperbolic tangent excitation function tansig,
and
two sets of weight parameters, P, for 6 neurons eachmean(k)And Pmean(k+1)And the predicted value of the average power of the two continuous time periods before and after the input of the input layer is obtained.
Optionally, the hidden layer and the output layer are connected by a linear function purelin, and a function expression of the output layer is as follows:
in the above formula, tanh represents a linear function purelin,
the six weight parameters are respectively connected between the output layer and the 6 neurons of the hidden layer.
Optionally, before the step 2), a step of performing offline training of the single hidden layer BP network in advance is further included:
s1) generating training data and dividing the training data into a training data set and a verification data set, wherein the training data set and the verification data set comprise average power predicted values of two continuous time periods before and after normalization processing and corresponding labels of energy storage state of charge reference values of the energy storage device;
s2) selecting a Levenberg-Marquardt back propagation optimization algorithm, training a single hidden layer BP network by adopting a training data set, and adjusting weight parameters of a hidden layer and an output layer by adopting a selection quantity gradient descent learning function Learndm;
s3) verifying the single hidden layer BP network of the training of the current round by adopting a verification data set, judging that the training is finished if the error of the single hidden layer BP network is less than a preset threshold value, saving the weight parameters of the hidden layer and the output layer as the finally obtained training result, and otherwise, skipping to execute the step S1) to continue the training.
Optionally, step 2) is followed by: acquiring an actual value of an energy storage state of charge of the energy storage device and a difference between the actual value and a reference value of the energy storage state of charge of the energy storage device to obtain an energy storage state of charge difference value delta SOC; and taking the energy storage state of charge difference value delta SOC and the current charge-discharge state of the battery energy storage system as the input of a fuzzy controller, and obtaining the current smoothing time constant T by the fuzzy controller according to the energy storage state of charge difference value delta SOC and the charge-discharge state by using a fuzzy rule.
Optionally, obtaining the current smoothing time constant T is followed by outputting the power P output by the wind farmGObtaining a wind power grid-connected power reference value P through a first-order low-pass filterTStep (c), when wind power plant outputs power PGHigher than the wind power grid-connected power reference value PTStoring excess energy in an energy storage device; when wind power plant outputs power PGLower than the wind power grid-connected power reference value PTWhen the power grid is connected with the power supply, the energy stored in the energy storage device is released to provide power support for the power grid; and the functional expression of the transfer function of the first-order low-pass filter is as follows:
in the above equation, T is a smoothing time constant.
In addition, the invention also provides an energy storage state of charge reference value prediction system of the wind power energy storage hybrid system, which comprises a microprocessor and a memory which are connected with each other, wherein the microprocessor is programmed or configured to execute the steps of the energy storage state of charge reference value prediction method of the wind power energy storage hybrid system.
Furthermore, the present invention also provides a computer readable storage medium having stored therein a computer program for execution by a computer device to implement the steps of the method for predicting an energy storage state of charge reference value of a wind energy and energy storage hybrid system.
Compared with the prior art, the invention has the following advantages: the method comprises the steps of obtaining the average power predicted value of the wind power energy storage hybrid system in the two nearest continuous time periods; the average power predicted values of the two continuous time periods before and after are input into a machine learning network model which is trained in advance and has two inputs and one output, so that an energy storage charge state reference value of an energy storage device in the wind power energy storage hybrid system is obtained. According to the method, the machine learning network model with two inputs and one output is used for predicting the energy storage state of charge reference value of the energy storage device based on the average power predicted value of two continuous time periods and the mapping relation between the energy storage state of charge reference values of the energy storage device, so that the prediction accuracy of the energy storage state of charge reference value of the energy storage device can be improved through a large amount of data sample training, the optimal smoothing effect of the power fluctuation of a wind power plant can be realized on the premise that the power and capacity of a battery energy storage system are configured to be certain, and the battery energy storage system can avoid the occurrence of the condition of over-charging or deep discharging.
Drawings
FIG. 1 is a schematic diagram of a basic flow of a method according to an embodiment of the present invention.
Fig. 2 is a schematic diagram of the division of the input signal of the machine learning network model according to the embodiment of the present invention.
Fig. 3 is a schematic structural diagram of a single hidden layer BP network according to an embodiment of the present invention.
Fig. 4 is a schematic structural diagram of a wind power and energy storage hybrid system according to an embodiment of the present invention.
Fig. 5 is a block diagram of a basic smooth control strategy of the battery energy storage system in the embodiment of the invention.
Fig. 6 is a schematic diagram of a grid-connected power curve of the wind power energy storage system in the embodiment of the invention.
Fig. 7 is a schematic diagram of an output power curve of the energy storage system according to the embodiment of the invention.
Fig. 8 is a schematic diagram of a state of charge curve of the energy storage system according to the embodiment of the invention.
FIG. 9 is a schematic block diagram of a fuzzy-neural network control for forming a smoothing time constant T in an embodiment of the present invention.
FIG. 10 is a control rule curve of the neural network controller according to an embodiment of the present invention.
Fig. 11 is a characteristic of a variation of the grid-connected power reference value PT with the smoothing time constant T according to the embodiment of the present invention.
FIG. 12 is a predicted value of the short-term output power of the wind farm in the embodiment of the invention.
FIG. 13 is an average value of the short-term output power of the wind farm in the embodiment of the present invention.
Fig. 14 is a state of charge reference value of the battery energy storage system according to the embodiment of the invention.
Fig. 15 is a simulation waveform when the fuzzy-neural network control strategy in the method of the present embodiment is adopted. The method comprises the following steps of (a) obtaining grid-connected power of a wind power plant output power/wind power energy storage hybrid system, (b) obtaining a smoothing time constant, (c) obtaining battery energy storage system output power, and (d) obtaining a battery energy storage system charge state monitoring value.
Fig. 16 is a simulation waveform of a system when a conventional control strategy is adopted, where (a) is grid-connected power of a wind power energy storage hybrid system, (b) is output power of a battery energy storage system, and (c) is a charge state monitoring value of the battery energy storage system.
Fig. 17 shows a simulation waveform of the system when the conventional control strategy is adopted and T is 150 s. Wherein, (a) is the grid-connected power of the wind power plant output power/wind power energy storage hybrid system, and (b) is the grid-connected power PTOutput waveforms when the fuzzy-neural network control is adopted and the traditional control is adopted, (c) is output power of the battery energy storage system, and (d) is a charge state monitoring value of the battery energy storage system.
Detailed Description
As shown in fig. 1, the method for predicting the energy storage state of charge reference value of the wind power energy storage hybrid system in the embodiment includes:
1) acquiring the average power predicted value of the wind power energy storage hybrid system in the latest two continuous time periods;
2) and inputting the average power predicted values of the two continuous time periods into a machine learning network model with two inputs and one output, which completes training in advance, so as to obtain the energy storage state of charge reference value of the energy storage device in the wind power energy storage hybrid system, wherein the machine learning network model with two inputs and one output is trained in advance to establish the mapping relation between the average power predicted values of the two continuous time periods and the energy storage state of charge reference value of the energy storage device.
In the embodiment, the short-term predicted power of the wind farm is divided into equal time intervals, and the average power of each time interval is obtained and used as an input signal of the machine learning network model, that is, the predicted average power value of two nearest consecutive time intervals, as shown in fig. 2.
As an alternative implementation, the two-input-one-output machine learning network model in this embodiment is a two-input-one-output single hidden layer BP network. As shown in FIG. 3, the single hidden layer BP network comprises an input layer, a hidden layer and an output layer, wherein the input layer comprises two neurons for inputting the average power predicted values of two continuous time periods before and after the input layer, and the hidden layer comprises 6 neurons for outputting X1～X6And the output layer is used for outputting the obtained energy storage charge state reference value of the energy storage device.
In this embodiment, the input layer and the hidden layer are connected by a hyperbolic tangent excitation function tansig, and the function expression of the hidden layer is as follows:
in the above formula, tanh represents a hyperbolic tangent excitation function tansig,
and
two sets of weight parameters, P, for 6 neurons eachmean(k)And Pmean(k+1)And the predicted value of the average power of the two continuous time periods before and after the input of the input layer is obtained.
In this embodiment, the hidden layer and the output layer are connected by a linear function purelin, and the function expression of the output layer is as follows:
in the above formula, tanh represents a linear function purelin,
the six weight parameters are respectively connected between the output layer and the 6 neurons of the hidden layer.
In this embodiment, step 2) further includes, before performing offline training of the single hidden layer BP network, a step of:
s1) generating training data and dividing the training data into a training data set and a verification data set, wherein the training data set and the verification data set comprise average power predicted values of two continuous time periods before and after normalization processing and corresponding labels of energy storage state of charge reference values of the energy storage device;
s2) selecting a Levenberg-Marquardt back propagation optimization algorithm, training a single hidden layer BP network by adopting a training data set, and adjusting weight parameters of a hidden layer and an output layer by adopting a selection quantity gradient descent learning function Learndm;
s3) verifying the single hidden layer BP network of the training of the current round by adopting a verification data set, judging that the training is finished if the error of the single hidden layer BP network is less than a preset threshold value, saving the weight parameters of the hidden layer and the output layer as the finally obtained training result, and otherwise, skipping to execute the step S1) to continue the training.
In this embodiment, step 2) further includes: acquiring an actual value of an energy storage state of charge of the energy storage device and a difference between the actual value and a reference value of the energy storage state of charge of the energy storage device to obtain an energy storage state of charge difference value delta SOC; and taking the energy storage state of charge difference value delta SOC and the current charge-discharge state of the battery energy storage system as the input of a fuzzy controller, and obtaining the current smoothing time constant T by the fuzzy controller according to the energy storage state of charge difference value delta SOC and the charge-discharge state by using a fuzzy rule. The fuzzy controller is used for giving a proper smooth time constant value according to the charge state and the charge-discharge state of the battery energy storage system, so that the wind power grid-connected power reference value is dynamically adjusted in real time, and the charge state of the battery energy storage system is controlled in real time. When the state of charge of the battery energy storage system is higher, if the battery energy storage system is in the state of charge, the smoothing time constant is reduced, and the tracking speed of the grid-connected power reference value to the output power of the wind power plant is increased, so that the magnitude of the charging power of the battery energy storage system is relatively reduced, the rising speed of the state of charge of the battery energy storage system is slowed down, and the condition of overcharge of the battery energy storage system is prevented; if the grid-connected power reference value is in the discharging state, the time constant is increased, the tracking speed of the grid-connected power reference value to the output power of the wind power plant is slowed, the discharging power of the battery energy storage system is relatively increased, the reduction of the charge state of the battery energy storage system is accelerated, and the battery energy storage system is changed towards the proper charge state. Vice versa, when the state of charge of the battery energy storage system is low, if the battery energy storage system is in the state of charge, the time constant is increased, and the tracking speed of the grid-connected power reference value to the output power of the wind power plant is slowed, so that the charging power of the battery energy storage system is relatively increased, the rise of the state of charge of the battery energy storage system is accelerated, and the battery energy storage system is changed towards the proper state of charge; if the grid-connected power reference value is in the discharging state, the time constant is reduced, the tracking speed of the grid-connected power reference value to the output power of the wind power plant is increased, and therefore the discharging power of the battery energy storage system is relatively reduced, the reduction speed of the charge state of the battery energy storage system is reduced, and the deep discharging condition of the battery energy storage system is prevented.
In this embodiment, according to the above characteristics of the battery energy storage system control system, the fuzzy controller adopts a two-input-single-output two-dimensional structure, and the input and output of the fuzzy controller include: input 1: the reference value of the state of charge output by the neural network controller is SOC, the real-time state of charge of the battery energy storage system is SOC, the deviation delta SOC of the state of charge is SOC-SOC, the linguistic variable is E2, and the basic discourse domain is [ -100%, 100%]The ambiguity domain is { -3, -2, -1,0, +1, +2, +3}, and the corresponding ambiguity subset is { NB, NM,NS, ZO, PS, PM and PB, which respectively represent that the state of charge of the current battery energy storage system is { very low, moderate, very high } relative to a set value. Input 2: and taking the language variable of the charge-discharge state of the battery energy storage system as E3. N represents that the battery energy storage system is in a discharging state, and P represents that the battery energy storage system is in a charging state. The charging and discharging state of the battery energy storage system can be output by the wind power plantGPower P connected to actual gridTIs determined if PG＜PTIndicating that the energy storage system is in a discharged state if PG＞PTIndicating that the energy storage system is in a state of charge. And (3) outputting: a smoothing time constant T with a basic universe of discourse [0s,3000 s%]The ambiguity domain is {0,1,2,3,4,5,6}, the corresponding ambiguity subset is { EL, VL, RL, ZO, RB, VB, EB }, and the output time constant is { minimum, moderate, large, maximum }, respectively. The input and output membership functions of the fuzzy controller adopt Gaussian membership functions with strong sensitivity, and the defuzzification method adopts a gravity center method. The control rule of the fuzzy controller is given according to the relationship between the deviation of the state of charge and the smooth time constant output corresponding to the charge and discharge state of the battery energy storage system, which is described above, and is shown in table 1.
Table 1: fuzzy controller rule table.
The fuzzy rule of table 1 fully reflects the variation characteristics of the smoothing time constant T in different state of charge deviations Δ SOC and in their charge-discharge states. The following 2 rules are selected from Table 1 and described.
Rule 1: IF E2 is PB and E3 is P, THEN T is EL;
rule 2: IF E2 is NB and E3 is P, THEN T is EB.
In this embodiment, obtaining the current smoothing time constant T further includes outputting the power P output by the wind farmGObtaining a wind power grid-connected power reference value P through a first-order low-pass filterTStep (c), when wind power plant outputs power PGHigher than the wind power grid-connected power reference value PTStoring excess energy in an energy storage device; when wind power plant outputs power PGLower than the wind power grid-connected power reference value PTWhen the power grid is connected with the power supply, the energy stored in the energy storage device is released to provide power support for the power grid; and the functional expression of the transfer function of the first-order low-pass filter is as follows:
in the above equation, T is a smoothing time constant.
As shown in fig. 4, the wind power storage hybrid system of the present embodiment incorporates a battery energy storage system between the wind farm outlet and the power grid, and reasonably controls the output power P of the battery energy storage systemFTo smooth the wind farm output power PGThe grid-connected power P of the wind power energy storage hybrid system is enabled to be achievedTA smoother output is realized; when wind power plant outputs power PGHigher than the wind power grid-connected power reference value PTWhen, i.e. PG>PTWhen the energy storage device is in operation, the energy storage device can store the redundant energy; (ii) a When wind power plant outputs power PGLower than the wind power grid-connected power reference value PTWhen, i.e. PG<PTThe energy stored in the energy storage device can be released to provide power support for the power grid.
Fig. 5 is a block diagram of a basic smooth control strategy of a battery energy storage system in the wind and electricity energy storage composite system in the embodiment. The grid-connected power reference value PT is obtained by filtering the wind farm output power PG through a first-order low-pass filter, and the time constant T of the first-order low-pass filter is referred to as a smoothing time constant T. Fig. 6 to 8 are power of the wind power energy storage system and a state of charge curve of the energy storage system in the embodiment of the present invention, and it can be known from fig. 6 to 8 that: the smaller the smoothing time constant T is, the faster the tracking speed of grid-connected power PT on output power PG of the wind power plant is, the lower the requirement of the maximum instantaneous power and capacity required by the battery energy storage system is, but the poorer the smoothing effect on power fluctuation of the wind power plant is. On the contrary, the larger the smoothing time constant T is, the better the smoothing effect of the battery energy storage system on the wind power fluctuation is, but at the same time, the battery energy storage system is also required to have larger power output capability and be configured with larger energy storage capacity. According to the relationship between the smoothing time constant T and the power fluctuation effect of the smooth wind power plant and the configuration relationship between the smoothing time constant and the power and capacity of the battery energy storage system, the fact that the fixed smoothing time constant cannot meet the requirements of good smoothing effect on the power fluctuation of the wind power plant and low configuration on the power and capacity of the battery energy storage system at the same time can be obtained.
In this embodiment, the main execution body of the machine learning network model is a neural network controller, and the neural network controller is used for giving a suitable state of charge reference value SOC of the energy storage device in a previous period according to predicted values of average output power of the wind farm in the previous and next periods. When the predicted value of the average power of the wind power plant in the later period is increased greatly compared with that in the earlier period, the reference value of the state of charge of the energy storage device in the earlier period is reduced, the energy storage device is enabled to be converted towards the lower state of charge in the earlier period, and a space with larger charging capacity possibly required in the later period is provided, so that the power fluctuation of the wind power plant can be smoothed to the maximum extent by the energy storage device in the later period. On the contrary, when the predicted value of the average output power in the later period of the wind power plant is reduced greatly compared with that in the earlier period, the reference value of the state of charge of the energy storage device in the earlier period is increased, so that the energy storage device is converted towards the higher state of charge in the earlier period, a larger discharge capacity space which is possibly needed in the later period is provided, and the later period is realized to the maximum extentAnd smooth output of grid-connected power. Fig. 9 is a schematic block diagram of the fuzzy-neural network control for forming the smoothing time constant T in the present embodiment. Firstly, predicting value P of average power of two periods before and after wind power plantmean(k)、Pmean(k+1)Obtaining a reference value SOC of the state of charge of the energy storage system in the previous period through a neural network control algorithm as the input of a neural network controller; and then, the deviation of the real-time SOC of the battery energy storage system from the reference value SOC and the current charge-discharge state of the battery energy storage system are used as the input of a fuzzy controller, and the size of the smoothing time constant T is dynamically adjusted in real time by using a fuzzy rule according to the deviation degree and the charge-discharge state. Therefore, the optimal smoothing effect of the power fluctuation of the wind power plant is achieved on the premise that the power and capacity of the battery energy storage system are configured to be certain, and meanwhile, the battery energy storage system can avoid the situation of overcharge or deep discharge.
In this embodiment, a Matlab neural network toolbox is specifically used to establish the single hidden layer BP network shown in fig. 3. The training function selects Levenberg-Marquardt back propagation optimization algorithm (Trainlm), the number of hidden layer neurons is 6, and the weight regulation rule selects a gradient descent learning function (Learndm). According to the corresponding relation between the average power predicted value and the state of charge reference value of the wind turbine generator in the two periods, 9 groups of typical input and target data can be determined to be used as training samples. In order to correct the network error performance and improve the generalization capability of the network, each training sample is normalized before training. The training error variation curve based on the neural network is shown in fig. 9, the training error is less than 0.0001 during the training of 160 steps, and the training result is shown in table 2. The offline trained single hidden layer BP network is used for real-time control of the wind power energy storage system online, and the control rule of the single hidden layer BP network is shown in FIG. 10.
Table 2: and (5) training results of the single hidden layer BP network.
FIG. 11 shows the power P output from a wind farmGWhen the power is 30MWGrid-connected power reference value PTWaveform plot of variation with smoothing time constant T. From the foregoing, the smaller the value of the smoothing time constant T is, the smaller the grid-connected power reference value P isTOutput power P of wind power plantGThe faster the tracking speed of (2) and the slower the other way around.
In order to verify the effectiveness of the proposed control strategy, the wind power and energy storage hybrid system in the embodiment is subjected to simulation analysis by using MATLAB/SIMULINK. The specific simulation parameters are as follows: rated output power of the wind power plant is 30MW, rated capacity of the battery energy storage system is 15MW/2.5MW & h, and grid-connected power P of the wind power systemTThe initial value at time 0s is 15 MW; the initial state of charge of the battery energy storage system at the time of 0s is 85%; and assuming that the normal working range of the state of charge of the battery energy storage system is 10% -100%; if the range is exceeded, the battery energy storage system stops working. The smoothing time constant T varies from 0s to 3000 s. The following are simulation results and analysis of the system using the fuzzy-neural network control strategy and using the conventional control strategy.
Fig. 12 is a predicted value of the short-term output power of the wind farm in 0min to 60min, and assuming that the actual output power of the wind farm is consistent with the predicted value, the average value of the output power of 10 min/time period is as shown in fig. 13, and the state of charge reference value SOC of the battery energy storage system corresponding to each time period can be obtained according to the neural network control algorithm, as shown in fig. 14. The average power of the wind power plant is 9.78MW at 0min-10min, the average power of the wind power plant is 22.08MW at 10min-20min, and the relative increase is 12.3 MW. The output of the reference value of the state of charge of the battery energy storage system is 23.14% in the period of 0min to 10min correspondingly, so that the battery energy storage system is controlled to be converted to a lower state of charge in the period of 0min to 10min, and a possibly required larger charging capacity space is provided for 10min to 20 min. The average power of the wind turbine generator set in 20min-30min is 12.75MW, and is relatively reduced by 9.33MW compared with the average power of 10min-20 min. The corresponding output of the reference value of the state of charge of the battery energy storage system in the period of 10min to 20min is 72.06%, so that the battery energy storage system is controlled to be converted to a higher state of charge in the period of 10min to 20min, and a possibly required larger discharge capacity space is provided for 20min to 30 min. When the average power change of the two periods is not large, the output of the reference value of the state of charge of the battery energy storage system is about 55%.
When the battery energy storage system adopts the fuzzy-neural network control strategy proposed by the method of the embodiment, the simulation result is shown in fig. 15. When the conventional control strategy is adopted by the battery energy storage system, the simulation results are shown in fig. 16 and 17. Generally, the larger the smoothing time constant value is, the better the smoothing effect on the power fluctuation of the wind farm is, but at the same time, a battery energy storage system with a larger capacity needs to be configured, which is not favorable for economy. Fig. 16 is a simulation result of the battery energy storage system under the conventional control strategy when the smoothing time constant is respectively 200s, 700s, 1100 s, 1700s and 3000 s. As can be seen from fig. 16, after the battery energy storage system is operated for a period of time, the state of charge of the battery energy storage system will reach 100%, and then the battery energy storage system will be in an overcharged state for a period of time. As shown in fig. 17, when the smoothing time constant T is 150s, although it can be ensured that the battery energy storage system does not have the condition of overcharge or deep discharge in the simulation period, the smoothing effect on the power fluctuation of the wind farm is not as good as that when the fuzzy-neural network control strategy is adopted, as shown in sub-graph (b) in fig. 17. By contrast, when the battery energy storage system adopts the fuzzy-neural network control strategy proposed by the method of the embodiment, and when the state of charge of the battery energy storage system is large, if the battery energy storage system is in the state of charge, the filtering time constant is reduced, the charging power is reduced, the increasing speed of the state of charge is slowed down, and the condition of overcharge is prevented; if the wind power station is in the discharging state, the filtering time constant is increased, the discharging power of the wind power station is increased, the reduction of the charging state of the wind power station is accelerated, the wind power station is enabled to be converted to the appropriate charging state, and meanwhile, a good smoothing effect can be achieved on the power fluctuation of the wind power station. Vice versa, the simulation result well illustrates the effectiveness and correctness of the control method provided by the embodiment.
In addition, the present embodiment further provides an energy storage state of charge reference value prediction system of a wind power energy storage hybrid system, which includes a microprocessor and a memory connected to each other, where the microprocessor is programmed or configured to execute the steps of the energy storage state of charge reference value prediction method of the wind power energy storage hybrid system.
In addition, the present embodiment also provides a computer readable storage medium, in which a computer program is stored, where the computer program is used to be executed by a computer device to implement the steps of the energy storage state of charge reference value prediction method of the wind power and energy storage hybrid system.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-readable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein. The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks. These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks. These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The above description is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments, and all technical solutions belonging to the idea of the present invention belong to the protection scope of the present invention. It should be noted that modifications and embellishments within the scope of the invention may occur to those skilled in the art without departing from the principle of the invention, and are considered to be within the scope of the invention.
Claims (10)
1. The method for predicting the energy storage state of charge reference value of the wind power energy storage hybrid system is characterized by comprising the following steps of:
1) acquiring the average power predicted value of the wind power energy storage hybrid system in the latest two continuous time periods;
2) and inputting the average power predicted values of the two continuous time periods into a machine learning network model with two inputs and one output, which completes training in advance, so as to obtain the energy storage state of charge reference value of the energy storage device in the wind power energy storage hybrid system, wherein the machine learning network model with two inputs and one output is trained in advance to establish the mapping relation between the average power predicted values of the two continuous time periods and the energy storage state of charge reference value of the energy storage device.
2. The method for predicting the energy storage state of charge reference value of the wind and electricity energy storage hybrid system according to claim 1, wherein the two-input and one-output machine learning network model is a two-input and one-output single hidden layer BP network.
3. The method for predicting the energy storage state of charge reference value of the wind and electricity energy storage hybrid system according to claim 2, wherein the single hidden layer BP network comprises an input layer, a hidden layer and an output layer, wherein the input layer comprises two neurons for inputting the average work of two continuous time periods before and after the input layerRate prediction value, implicit layer comprising 6 neurons for outputting X1～X6And the output layer is used for outputting the obtained energy storage charge state reference value of the energy storage device.
4. The method for predicting the energy storage state of charge reference value of the wind power and energy storage hybrid system according to claim 3, wherein the input layer and the hidden layer are connected by adopting a hyperbolic tangent excitation function tansig, and the function expression of the hidden layer is as follows:
in the above formula, tanh represents a hyperbolic tangent excitation function tansig,
and
two sets of weight parameters, P, for 6 neurons eachmean(k)And Pmean(k+1)And the predicted value of the average power of the two continuous time periods before and after the input of the input layer is obtained.
5. The method for predicting the energy storage state of charge reference value of the wind and electricity energy storage hybrid system according to claim 4, wherein the hidden layer and the output layer are connected by a linear function purelin, and the function expression of the output layer is as follows:
in the above formula, tanh represents a linear function purelin,
six weight parameters respectively connected between the output layer and 6 neurons of the hidden layer。
6. The method for predicting the energy storage state of charge reference value of the wind and electricity energy storage hybrid system according to claim 5, wherein the step 2) is preceded by a step of performing offline training of the single hidden layer BP network:
s1) generating training data and dividing the training data into a training data set and a verification data set, wherein the training data set and the verification data set comprise average power predicted values of two continuous time periods before and after normalization processing and corresponding labels of energy storage state of charge reference values of the energy storage device;
s2) selecting a Levenberg-Marquardt back propagation optimization algorithm, training a single hidden layer BP network by adopting a training data set, and adjusting weight parameters of a hidden layer and an output layer by adopting a selection quantity gradient descent learning function Learndm;
s3) verifying the single hidden layer BP network of the training of the current round by adopting a verification data set, judging that the training is finished if the error of the single hidden layer BP network is less than a preset threshold value, saving the weight parameters of the hidden layer and the output layer as the finally obtained training result, and otherwise, skipping to execute the step S1) to continue the training.
7. The method for predicting the energy storage state of charge reference value of the wind and electricity energy storage hybrid system according to claim 1, wherein the step 2) is followed by the steps of: acquiring an actual value of an energy storage state of charge of the energy storage device and a difference between the actual value and a reference value of the energy storage state of charge of the energy storage device to obtain an energy storage state of charge difference value delta SOC; and taking the energy storage state of charge difference value delta SOC and the current charge-discharge state of the battery energy storage system as the input of a fuzzy controller, and obtaining the current smoothing time constant T by the fuzzy controller according to the energy storage state of charge difference value delta SOC and the charge-discharge state by using a fuzzy rule.
8. The method for predicting the energy storage state of charge reference value of the wind and electricity energy storage hybrid system according to claim 7, wherein the step of obtaining the current smoothing time constant T further comprises the step of predicting the power P output by a wind farmGObtained by a first-order low-pass filterWind power integration power reference value PTStep (c), when wind power plant outputs power PGHigher than the wind power grid-connected power reference value PTStoring excess energy in an energy storage device; when wind power plant outputs power PGLower than the wind power grid-connected power reference value PTWhen the power grid is connected with the power supply, the energy stored in the energy storage device is released to provide power support for the power grid; and the functional expression of the transfer function of the first-order low-pass filter is as follows:
in the above equation, T is a smoothing time constant.
9. An energy storage state of charge reference value prediction system of a wind power energy storage hybrid system, comprising a microprocessor and a memory which are connected with each other, characterized in that the microprocessor is programmed or configured to execute the steps of the energy storage state of charge reference value prediction method of the wind power energy storage hybrid system according to any one of claims 1 to 8.
10. A computer-readable storage medium, in which a computer program is stored, wherein the computer program is used for being executed by a computer device to implement the steps of the method for predicting the energy storage state of charge reference value of a wind power and energy storage hybrid system according to any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111415712.6A CN114123279B (en) | 2021-11-25 | 2021-11-25 | Method and system for predicting energy storage state of charge reference value of wind power energy storage hybrid system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111415712.6A CN114123279B (en) | 2021-11-25 | 2021-11-25 | Method and system for predicting energy storage state of charge reference value of wind power energy storage hybrid system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114123279A true CN114123279A (en) | 2022-03-01 |
| CN114123279B CN114123279B (en) | 2023-11-21 |
Family
ID=80373445
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111415712.6A Active CN114123279B (en) | 2021-11-25 | 2021-11-25 | Method and system for predicting energy storage state of charge reference value of wind power energy storage hybrid system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114123279B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102231532A (en) * | 2011-06-28 | 2011-11-02 | 青岛派克能源有限公司 | Liquid flow battery energy storage paralleling apparatus and method |
| CN109327031A (en) * | 2018-09-30 | 2019-02-12 | 国网湖南省电力有限公司 | Directly driven wind-powered multi-computer system power association control method and system based on battery energy storage |
| CN111342471A (en) * | 2020-03-02 | 2020-06-26 | 华北电力大学 | Machine learning-based family obstetrician and consumer power optimization management method |
| CN113112099A (en) * | 2021-05-14 | 2021-07-13 | 国网河北省电力有限公司经济技术研究院 | Power grid daily electric quantity prediction model training method and power grid daily electric quantity prediction method |
-
2021
- 2021-11-25 CN CN202111415712.6A patent/CN114123279B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102231532A (en) * | 2011-06-28 | 2011-11-02 | 青岛派克能源有限公司 | Liquid flow battery energy storage paralleling apparatus and method |
| CN109327031A (en) * | 2018-09-30 | 2019-02-12 | 国网湖南省电力有限公司 | Directly driven wind-powered multi-computer system power association control method and system based on battery energy storage |
| CN111342471A (en) * | 2020-03-02 | 2020-06-26 | 华北电力大学 | Machine learning-based family obstetrician and consumer power optimization management method |
| CN113112099A (en) * | 2021-05-14 | 2021-07-13 | 国网河北省电力有限公司经济技术研究院 | Power grid daily electric quantity prediction model training method and power grid daily electric quantity prediction method |
Non-Patent Citations (1)
| Title |
|---|
| 崔建双主著: "《25个经典的元启发式算法-从设计到matlab实现》", 企业管理出版社, pages: 217 - 228 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114123279B (en) | 2023-11-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105846461B (en) | Control method and system for large-scale energy storage power station self-adaptive dynamic planning | |
| WO2021164112A1 (en) | Frequency control method and system during using wind farm as black-start power source by means of optimal configuration of energy storage | |
| Khooban et al. | A new load frequency control strategy for micro-grids with considering electrical vehicles | |
| CN104993522B (en) | A kind of active distribution network Multiple Time Scales coordination optimization dispatching method based on MPC | |
| CN103956758B (en) | Energy storage SOC optimal control method in a kind of wind storage system | |
| CN112952862B (en) | Hybrid energy storage frequency division coordination controller for stabilizing wind power fluctuation and implementation method | |
| CN107394798B (en) | Electric automobile and generator set coordinated frequency control method containing time-varying time lag | |
| CN105896575B (en) | Hundred megawatt energy storage power control method and system based on self-adaptive dynamic programming | |
| CN102738817B (en) | Method for carrying out self-adaptive smoothing treatment on wind power by battery energy storage system | |
| CN105406496A (en) | Isolated microgrid frequency modulation control method based on measured frequency response identification | |
| CN115102239A (en) | Energy storage power station primary frequency modulation control method and system considering SOC balance | |
| CN106505604A (en) | The photovoltaic energy storage cooperation unit optimization collocation method of access area power distribution network | |
| CN111697597A (en) | Fire storage combined AGC frequency modulation control method based on particle swarm optimization | |
| CN116667389A (en) | Novel wind-storage combined prediction correction control method for inertia lifting of power system | |
| CN113113916B (en) | Multi-target energy regulation and control method in ship comprehensive power system | |
| Yin et al. | Hybrid multi-agent emotional deep Q network for generation control of multi-area integrated energy systems | |
| CN111654054A (en) | Control method for stabilizing short-term wind power fluctuation based on Adaptive Neural Network (ANN) during energy storage | |
| CN112531735A (en) | Power distribution method and device of automatic power generation control system based on machine learning | |
| CN109327031A (en) | Directly driven wind-powered multi-computer system power association control method and system based on battery energy storage | |
| CN115102228A (en) | Multi-target coordination frequency optimization method and device for wind power plant containing flywheel energy storage | |
| CN114123279B (en) | Method and system for predicting energy storage state of charge reference value of wind power energy storage hybrid system | |
| CN112838598A (en) | Optimization control strategy based on self-adaptive continuous tabu search algorithm | |
| CN111525596B (en) | Double-battery fluctuation out-of-limit optimization method in wind storage combined system | |
| CN116050461A (en) | Improved method for determining membership function of fuzzy control theory by using convolutional neural network | |
| CN110556868B (en) | Wind power access system energy storage charging and discharging control method considering voltage stability |
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 |