CN115296308A - Robust cooperative frequency modulation method considering energy storage charge state and adaptive inertia level - Google Patents

Robust cooperative frequency modulation method considering energy storage charge state and adaptive inertia level Download PDF

Info

Publication number
CN115296308A
CN115296308A CN202211226501.2A CN202211226501A CN115296308A CN 115296308 A CN115296308 A CN 115296308A CN 202211226501 A CN202211226501 A CN 202211226501A CN 115296308 A CN115296308 A CN 115296308A
Authority
CN
China
Prior art keywords
power
energy storage
frequency
output
inertia
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
Application number
CN202211226501.2A
Other languages
Chinese (zh)
Other versions
CN115296308B (en
Inventor
曾伟
窦晓波
熊俊杰
陈拓新
范瑞祥
余侃胜
赵伟哲
李佳
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
State Grid Corp of China SGCC
Southeast University
Electric Power Research Institute of State Grid Jiangxi Electric Power Co Ltd
Original Assignee
State Grid Corp of China SGCC
Southeast University
Electric Power Research Institute of State Grid Jiangxi Electric Power Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by State Grid Corp of China SGCC, Southeast University, Electric Power Research Institute of State Grid Jiangxi Electric Power Co Ltd filed Critical State Grid Corp of China SGCC
Priority to CN202211226501.2A priority Critical patent/CN115296308B/en
Publication of CN115296308A publication Critical patent/CN115296308A/en
Application granted granted Critical
Publication of CN115296308B publication Critical patent/CN115296308B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

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
    • H02J3/24—Arrangements for preventing or reducing oscillations of power in networks
    • H02J3/241—The oscillation concerning frequency
    • 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
    • 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
    • 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
    • H02J3/466—Scheduling the operation of the generators, e.g. connecting or disconnecting generators to meet a given demand
    • 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
    • H02J3/48—Controlling the sharing of the in-phase component
    • 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
    • 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
    • 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
    • H02J2300/40—Systems for supplying or distributing electric power characterised by decentralized, dispersed, or local generation wherein a plurality of decentralised, dispersed or local energy generation technologies are operated simultaneously

Landscapes

  • Engineering & Computer Science (AREA)
  • Power Engineering (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Evolutionary Computation (AREA)
  • General Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Health & Medical Sciences (AREA)
  • Control Of Eletrric Generators (AREA)

Abstract

The invention discloses a robust cooperative frequency modulation method considering an energy storage charge state and a self-adaptive inertia level, which comprises the following steps of: establishing an active-frequency response model of the energy storage unit participating in power grid frequency modulation, and establishing a collaborative control model after the wind, light, water and fire storage multi-station is accessed to the power system; designing a virtual inertia control link on the basis of a multi-field station cooperative frequency modulation state space model to realize the simulation of frequency closed-loop control inertia; establishing a multi-field station cooperative frequency modulation state space model considering a virtual inertia ring, and adopting
Figure 875981DEST_PATH_IMAGE001
Designing a closed-loop controller by a robust method; based on a deep neural network method, an inertia control coefficient self-adaptive adjusting method is established, and the frequency dynamic response performance under a multi-working-condition multi-disturbance scene is improved. The invention has the following effects: the method can be used for solving the problem of insufficient inertia level possibly existing when the distributed new energy high-proportion access power system frequency modulation service is carried out, and can provide support for participating in operation control of the power system through the energy storage system.

Description

Robust cooperative frequency modulation method considering energy storage charge state and adaptive inertia level
Technical Field
The invention belongs to the technical field of energy storage system cooperative operation and control, and particularly relates to a wind, light, water and fire storage robust cooperative frequency modulation method considering energy storage charge state and self-adaptive inertia level.
Background
With increasing concerns about energy shortages and environmental issues, distributed power generation has received much attention and has been rapidly developed in recent years. However, most distributed power supplies are connected to the main grid, in particular the inverter, through power electronics, the damping and inertia characteristics of which are considered negligible. Obviously, increasing the permeability of a distributed power supply threatens its safety and stability, especially in the event of fluctuations in electrical energy. Therefore, virtual synchronous inertia control techniques are proposed to improve the inertia of the power system. Currently, a great deal of research is carried out on a new energy scheduling method for wind energy, photovoltaic energy, hydropower and energy storage, which mostly starts from the uncertainty of new energy, but less considers the peak regulation capability of the new energy. Along with the permeability of the distributed power supply is continuously improved, the anti-interference capability of a source and a power grid is poorer and poorer, the safety and the stability of the distributed power supply are greatly threatened, and how to cooperatively control the frequency and consider the inertia levels of different types of units plays an important role in the reliable and stable operation of a system.
Based on this need, various new energy sources have been developed rapidly. However, due to the strong uncertainty of renewable power generation and the current renewable energy power prediction accuracy still cannot meet the operation requirement, the safety and quality of the grid frequency are seriously affected. The frequency modulation method with fixed virtual inertia and damping has some disadvantages, and its dynamic response capability needs to be further improved. To solve this problem, adaptive virtual synchronization parameters propose a deep neural network-based adjustment strategy. The method utilizes a good approximation capability network of a deep neural network as a nonlinear function, has strong learning capability and can adjust parameters on line in real time. With the help of the simulation, the validity of the proposed strategy was verified. The proposed control strategy has the advantages of small inertia and high recovery speed, and can effectively inhibit power oscillation, thereby overcoming the defects of the traditional fixed parameter frequency modulation.
Therefore, the scheduling method which can consider the cooperative frequency modulation capability of the fan, the photovoltaic, the hydroelectric power, the thermal power and the energy storage to ensure the economical efficiency and the reliability of the operation of the power system is very significant and solves the problems that the existing power grid scheduling cost is high, the peak-shaving pressure is high, the new energy consumption is small, and the frequency modulation cooperative system is unstable and not rapid. The key point for solving the problems is to establish a robust cooperative frequency modulation model which comprises an energy storage charge state and a self-adaptive inertia level, covers five types of energy sources of wind, light, water and fire storage and considers the inertia level of the system.
The similar technical scheme in the market at present is 'an active frequency modulation closed-loop control method of an energy storage system'. The method can realize the fine control of each energy storage subunit during the active frequency modulation action of the energy storage system, and form closed-loop control of starting the frequency modulation of the energy storage system, finishing the smooth exit of the frequency modulation of the energy storage system and automatically recovering the charge state of the energy storage system; meanwhile, the charge state of the energy storage battery can be ensured to be in a normal frequency modulation working range all the time, the situation that the electric quantity of the energy storage system rapidly reaches a protection value due to the fact that the daily deviation direction of the frequency of a power grid is unbalanced is avoided, the energy storage system can be safely quitted when the frequency is abnormal or the charge state of the energy storage battery is abnormal can be guaranteed, and the dynamic response capability of active frequency modulation of the energy storage system and the reliability of long-term operation are further improved.
However, most of the research results at present are focused on the problem of frequency stability under the traditional control mode, which results in poor dynamic characteristics of the frequency, and the research results have the following disadvantages: the research does not consider the improvement of the dynamic performance of the system when the power grid is subjected to cooperative frequency modulation under wind, light, water and fire storage; secondly, the above researches do not pay attention to the influence of the adaptive inertia level existing in the control process on the system frequency control.
Disclosure of Invention
The invention aims to provide a robust cooperative frequency modulation method considering the energy storage charge state and the self-adaptive inertia level, overcomes the defect that the frequency stability is kept under the condition that the current renewable energy is accessed into a power grid environment, considers the cooperative action of wind, light, water, fire and energy storage, and has obvious advantages in the aspects of ensuring the robustness of a system and resisting the uncertain disturbance of the system.
The invention adopts the following technical scheme: the robust cooperative frequency modulation method considering the energy storage charge state and the self-adaptive inertia level is characterized by comprising the following steps of:
step 1, establishing an active-frequency response model of an energy storage unit participating in power grid frequency modulation, and further establishing a relation model of the state of charge and the output power of the energy storage unit;
step 2, considering other facilities influencing the frequency in the power grid, and establishing other equipment models of the power grid except the energy storage unit;
step 3, constructing a multi-station cooperative frequency modulation state space model containing the wind, light, water and fire power storage system;
step 4, designing a virtual inertia control link on the basis of a multi-station cooperative frequency modulation state space model to realize the simulation of frequency closed-loop control inertia;
step 5, establishing a multi-station cooperative frequency modulation state space model considering the virtual inertia ring, and designing a closed-loop controller by adopting an H infinity robust method in consideration of frequent disturbance of renewable energy sources to realize closed-loop stable control of frequency;
and 6, establishing an inertia control coefficient self-adaptive adjusting method based on a deep neural network method, and realizing the improvement of the dynamic response performance of the frequency under the multi-working-condition multi-disturbance scene.
Further, in the step 1, an active-frequency response model of the energy storage unit participating in the frequency modulation of the power grid is established, and a relation model of the state of charge and the output power of the energy storage unit is further established:
Figure 660713DEST_PATH_IMAGE001
（1）
in the formula,. DELTA.P BESS_i For energy-storage power stationsiThe active power variation is output,β iBESS representing energy storage power stationiResponse time constant, Δh BESSi_sig For energy-storage power stationsiThe control signal is sent to the computer system,α BESSi_sig for energy-storage power stationsiPower signal division coefficient, ΔP ci Representing energy storage power stationiAn active power difference control signal is generated in response to the active power difference control signal,sis a differential operator.
Establishing a state estimation model for the energy storage unit:
Figure 904744DEST_PATH_IMAGE002
（2）
in the formula (I), the compound is shown in the specification,SOC BESS (T+Δt) RepresentT+ΔtThe charge capacity of the SOC unit at the time,SOC BESS (T) To representTCharge capacity, Δ, of the time-of-day SOC celltFor a given time interval it is possible to provide,i o the current is output for the energy storage battery,ta time variable is represented by a time variable,C cap BESS_ the ampere capacity of the energy storage battery is obtained.
The constraint conditions of the energy storage unit are as follows:
Figure 408537DEST_PATH_IMAGE003
（3）
in the formula (I), the compound is shown in the specification,SOC BESSmin represents the minimum charge capacity of the SOC,SOC BESSmax represents the maximum charge capacity of the SOC,SOC BESSi representing energy storage power stationiThe SOC charge capacity of (a) is,P i_min indicating a minimum output active power limit and,P i_max the maximum output active power limit is indicated,P i representing energy storage power stationiThe SOC outputs active power.
Before participating in frequency modulation, the respective charge states of the energy storage units are detected according to the unitsSOCDeviation and output power limit adjustment output:
Figure 506943DEST_PATH_IMAGE004
（4）
in the formula,. DELTA.P i Representing energy storage power stationiActive power difference, Δ, of the SOC outputu BESSi For energy-storage power stationsiControl command, Δu sigSOC A charge state adjusting signal of the energy storage power station,β BESS representing the response time constant of the energy storage plant.
Further, in the step 2, establishing models of other devices of the power grid except the energy storage unit respectively as follows:
hydroelectric and thermal power unit speed regulator model:
Figure 805201DEST_PATH_IMAGE005
（5）
in the formula,. DELTA.K gi For hydroelectric and thermal power generating unitsiThe position variation of the speed regulator regulating valve,β gi_speed for hydroelectric and thermal power generating unitsiTime constant of governor, Δu Gi_sig For hydroelectric and thermal power generating unitsiThe control signal is sent to the computer system,α ci for hydroelectric and thermal power generating unitsiThe power signal distribution coefficient is set to be,L ci for hydroelectric and thermal power unitsiThe coefficient of the droop is,
Figure 262159DEST_PATH_IMAGE006
is a regioniHydroelectric and thermal power generator unitiThe frequency of (d); deltaP ci For hydroelectric and thermal power generating unitsiThe active power variation.
Hydroelectric and thermal engine turbine models:
Figure 428698DEST_PATH_IMAGE007
（6）
in the formula,. DELTA.P mwi For hydroelectric and thermal power unitsiThe amount of change in the output power,β gi_turb for hydroelectric and thermal power generating unitsiThe turbine time constant.
Hydroelectric and thermal power machine secondary frequency modulation model:
Figure 256977DEST_PATH_IMAGE008
（7）
in the formula,. DELTA.P ACE_sig For secondary frequency-modulated signals, Δ, of the gridfAs the amount of change in the system frequency,K ci is the secondary frequency modulation coefficient.
The photovoltaic power station is adopted to actively reduce and participate in system frequency adjustment, the photovoltaic power station is equivalent to a first-order inertia link, and a dynamic response model is as follows:
Figure 991715DEST_PATH_IMAGE009
（8）
in the formula,. DELTA.P iPV Representing photovoltaic power stationsiThe amount of power change, Δ, of the outputu i_sigPv Representing photovoltaic power stationsiThe control signal is sent to the computer system,α iPV for photovoltaic power stationsiPower signal division factor, ΔP ci Representing photovoltaic power stationsiAn active power difference control signal is generated in response to the active power difference control signal,β iPV for photovoltaic power stationsiA response time constant.
The wind farm frequency response model may be expressed as:
Figure 538234DEST_PATH_IMAGE010
（9）
in the formula (I), the compound is shown in the specification,
Figure 649409DEST_PATH_IMAGE011
is a reference power for the wind farm,
Figure 597774DEST_PATH_IMAGE012
for the reference frequency of the wind farm,
Figure 503413DEST_PATH_IMAGE013
is the actual angular frequency of the wind farm,
Figure 268719DEST_PATH_IMAGE014
which is the wind speed,C p is the wind energy utilization coefficient and the speed ratio of the wind energy utilization coefficient to the wind turbine blade tip
Figure 918006DEST_PATH_IMAGE015
Pitch angle
Figure 720877DEST_PATH_IMAGE016
In connection with this, the first and second electrodes,
Figure 797418DEST_PATH_IMAGE017
in order to be the density of the air,Ris the radius of the area swept by the blade,J t is the inertia coefficient of the fan rotor and the generator.
Further, in step 3, the multi-field station cooperative frequency modulation state space model is:
Figure 52949DEST_PATH_IMAGE018
（10）
in the formula (I), the compound is shown in the specification,H sys in order to obtain the integral inertia of the regional power grid,H iSG generating set for water and fireiThe inertia of the rotor is reduced to zero,H jWF for wind-power generation unitsjThe inertia of the rotor is reduced to zero,H kPV for photovoltaic unitskThe inertia of the rotor is reduced to zero,H mBESS for energy-storage power stationsmThe inertia of (a);S sys 、S Gi 、S jWF 、S kPV 、S mBESS respectively a system, a hydroelectric power unit and a thermal power unitiWind turbine generator setjPhotovoltaic unitkEnergy storage power stationmThe distribution coefficient of the inertia of (a),n s ，n w ，n p ，n e the number of the water-fire generator sets, the wind turbine generators, the photovoltaic sets and the energy storage power stations in the region is respectively.
The frequency response model is:
Figure 505928DEST_PATH_IMAGE019
（11）
in the formula,. DELTA.f(k+1) Is composed ofkFrequency difference, Δ, between +1 time and targetf(k) Is composed ofkFrequency difference, delta, of time of day and targetP WF (k)、ΔP PV (k)、ΔP BESS (k)、ΔP D (k) Respectively a wind power plant, a photovoltaic power plant, an energy storage, water and electricity and a thermal power generating unitkThe active balance of the time of day,T T is a constant of proportionality that is,D f is the damping coefficient of the region or regions,P LN in order to be the amount of power disturbance,Has a function of the frequency response of the system,f n is the nominal frequency.
The overall state space equation for a regional power system can be expressed as:
Figure 163305DEST_PATH_IMAGE020
（12）
in the formula (I), the compound is shown in the specification,x(t) Is at leasttThe overall state vector of the time system is,Ais a matrix of the whole system, and the system,Bis an integral control matrix and is characterized by that,B ω is an overall disturbance matrix and is characterized in that,M(t) Is at the same timetThe time of day is the overall control vector of the system,ω(t) Is at the same timetAnd (5) disturbance vectors of the whole system at the moment.
Further, in step 4, a virtual inertia control link is designed for the problems of insufficient inertia and poor frequency dynamic response performance provided by the water and thermal power generating units. The method can simulate the inertia constant and the damping characteristic of the synchronous generator, thereby improving the robustness and the sensitivity of the system. The virtual inertial control element can be represented by the following equation:
Figure 410747DEST_PATH_IMAGE021
（13）
in the formula (I), the compound is shown in the specification,P ref (s) Is a Laplace transform function which is referred by the inversion unit and represents the system energy shortage obtained by temporary calculation, the output of the Laplace transform function is transmitted to the wind, light, water, fire and storage and other cooperative frequency modulation units in a signal form,H i andD i is a virtual rotor parameter, Δf(s) Is the system frequency variation deltafThe laplace transform function of (a).
The overall transfer function can be expressed as:
Figure 422084DEST_PATH_IMAGE022
（14）
in the formula (I), the compound is shown in the specification,R i andK i respectively, a virtual droop constant and a virtual integral gain.
Based on the above transfer function, the frequency segment taking into account the virtual inertia can be designed as follows:
Figure 413173DEST_PATH_IMAGE023
（15）
in the formula (I), the compound is shown in the specification,
Figure 190637DEST_PATH_IMAGE024
is the frequency variation of the system, ΔP 1 Active power regulation, delta, for regional wind, light, water and fire storageP 2 For uncontrolled disturbances, Δ, inside and outside the regionPFor regional out-bound active power output variations,J sys is a virtual inertia of the system and,
Figure 874559DEST_PATH_IMAGE025
in order to be the nominal frequency of the system,sin order to be the laplace transform operator,D sys is the damping coefficient of the system.
Further, in the step 5, on the basis of the virtual inertia link, a multi-station cooperative frequency modulation state space model considering the virtual inertia ring is established, and in consideration of frequent disturbance of renewable energy, a closed-loop controller is designed by adopting an H infinity robust methodM=KxTo ensure the system to be disturbed by the outsideiAnd the stability under the condition is improved, the influence of disturbance on a regulated output function of the system is reduced to the maximum extent, and the closed-loop stable control of the frequency can be realized. Corresponding closed loop system:
Figure 839104DEST_PATH_IMAGE026
（16）
in the formula (I), the compound is shown in the specification,Min order to be a matrix of the states of the controller,xinputting a state vector for a systemoutput(k) Is composed ofkThe output function of the time of day system,C 1 in order to output the matrix for the state,D 12 in order to perturb the output matrix,D 11 in order to control the output matrix,Kin order to achieve the gain,x(k) Is a systemkThe state variable that is input at a time is,i(k) Is composed ofkDisturbance variables of the time system;
the robust performance constraint inequality of the system is:
Figure 633884DEST_PATH_IMAGE027
（17）
wherein, the first and the second end of the pipe are connected with each other,Ais a matrix of parameters, and is,Gin the form of a matrix of parameters,WandXis a positive definite matrix to be solved, T represents the matrix transposition,
Figure 265854DEST_PATH_IMAGE028
a parameter matrix is designed for the controller,
Figure 589519DEST_PATH_IMAGE029
for the controller transfer function minimum rejection ratio, Iis a matrix of the units,Zthe output matrix is evaluated for the controller, C 2 representing a state parameter matrix.
Positive definite matrix if there is optimal solutionWAndXthe system is progressively stabilized and the gain parameter is controlled to beK=X(W) -1 I.e. byu(t)=X(W) -1 xFor co-frequency modulationHAnd an infinity controller. The linear matrix inequality is used for solving the control parameters, so that the solving complexity is reduced, and the optimal control parameters are ensured.
Further, in step 6, an inertia control coefficient adaptive adjustment method is established based on a deep neural network method.
By utilizing the neural network, the operation state can be adjusted according to different conditions and the virtual inertia system can be controlled in real time. The input data to the neural network are the regional frequency deviation and the rate of change of frequency, the input being converted to the two-dimensional mapping and pooling layers in the convolutional layer. And finally, a complete connection layer containing the output control signal and the system inertia parameter deviation. The convolutional layer contains a plurality of convolution kernels, and each element corresponds to a weight coefficient and a deviation value. The size of the convolution kernel is determined by the convolution kernel, and the description of the features of the normalization input data extracted by the convolution layer is as follows:
Figure 569589DEST_PATH_IMAGE030
（18）
in the formula (I), the compound is shown in the specification,O n j, the output of the convolutional layer is shown,fmapping for layers，
Figure 902481DEST_PATH_IMAGE031
Representing a selected input mapping set, and generating an output matrix with a determined size through a selected input with a certain size and convolution kernel operation, and adding a plurality of weights and offset parameters;U i n,-1 represents a parameterized spatial filter, the specific action of which is automatically learned from the data during training of the network,M n ij, which represents the input of the convolutional layer,b n j, in the case of a real number bias term,
Figure 388957DEST_PATH_IMAGE032
represents a cross product;
the function of the pooling layer is to select features and filter layers of information from the convolved output data, which is expressed as follows:
Figure 414682DEST_PATH_IMAGE033
（19）
in the formula (I), the compound is shown in the specification,Q n+1,j the water is output from the pool layer,gin order to pool the mapping, the mapping is performed,k n+1,j andr n+1,j respectively representing the multiplicative deviation and the additive deviation,
Figure 619399DEST_PATH_IMAGE034
representing a pooling function;
the filter layer is thus represented as:
Figure 615036DEST_PATH_IMAGE035
（20）
in the formula (I), the compound is shown in the specification,p、qrespectively representing the length and width of the filter layer; the pooling operation is to reduce the output of the upper layers, which makes the transformation of the predicted input data more robust,J i,j is the output of the filtering layer, and is the output of the filtering layer,S p,q in order to filter the parameters of the process,b i+p ,j+q--11 is a real bias term;
finally, the complete connection layer expands the characteristic diagram into vectors and the output value of each complete connection layer; passing to the output requires defining a loss function to accomplish this classification; the fully connected layer performs non-linear combination, and this process is described as follows:
Figure 956019DEST_PATH_IMAGE036
（21）
in the formula (I), the compound is shown in the specification,Y n represents the output of the fully-connected layer,fa non-linear function representing a fully connected layer,w n andp n respectively representing the weight and the deviation of the fully connected layer.
The invention has the technical effects that: by adopting the scheme, the method can be used for making up the defect that the traditional thermal power generating unit mainly participates in the current power system frequency modulation service, and can provide support for the energy storage system to participate in the operation control of the power system; compared with the existing energy storage system participating in the frequency modulation control of the power system, the method focuses on more precise analysis and expression of rapid recovery of the power grid frequency containing distributed power supply uncertainty, reasonably distributes the active output of the energy storage unit by utilizing the charge state balance design in the energy storage power station participating in the frequency modulation service of the power system, establishes the system robust performance index as the evaluation standard, and has obvious advantages in the aspects of ensuring the system robustness and resisting the system uncertainty disturbance; aiming at the processing of the self-adaptive inertia level, the closed-loop stable control of the system frequency and the promotion of the frequency dynamic response performance under the multi-working-condition multi-disturbance scene are realized by establishing a multi-station cooperative frequency modulation state space model considering the virtual inertia ring and considering the disturbance of renewable energy sources and the like.
Drawings
FIG. 1 is a block diagram of the overall model of the present invention.
FIG. 2 is a flow chart of the present invention.
FIG. 3 is a diagram illustrating the effect of improving the dynamic performance of the frequency according to the present invention.
Detailed Description
As shown in fig. 1-2, the present invention is a robust coordinated frequency modulation method that is operated and implemented in such a way as to account for energy storage state of charge and adaptive inertia level, and is characterized by comprising the following steps:
step 1, establishing an active-frequency response model of an energy storage unit participating in power grid frequency modulation, and further establishing a relation model of the state of charge and the output power of the energy storage unit;
step 2, considering other facilities influencing the frequency in the power grid, and establishing other equipment models of the power grid except the energy storage unit, such as: wind generating sets, photovoltaic power stations, hydropower stations, traditional thermal power generating units, loads and the like;
step 3, constructing a multi-station cooperative frequency modulation state space model containing a wind, light, water and fire power storage system;
step 4, designing a virtual inertia control link to realize the simulation of frequency closed-loop control inertia on the basis of a multi-field station cooperative frequency modulation state space model;
step 5, establishing a multi-station cooperative frequency modulation state space model considering the virtual inertia ring, and designing a closed-loop controller by adopting an H infinity robust method in consideration of frequent disturbance of renewable energy sources to realize closed-loop stable control of frequency;
and 6, establishing an inertia control coefficient self-adaptive adjusting method based on a deep neural network method, and realizing the improvement of the dynamic response performance of the frequency under the multi-working-condition multi-disturbance scene.
In the step 1, an active-frequency response model of the energy storage unit participating in power grid frequency modulation is established, and a relation model of the state of charge and the output power of the energy storage unit is further established:
Figure 152645DEST_PATH_IMAGE037
（1）
in the formula,. DELTA.P BESS_i For energy-storage power stationsiThe active power variation is output,β iBESS representing energy storage power stationiResponse time constant, Δh BESSi_sig For energy-storage power stationsiThe control signal is sent to the computer system,α BESSi_sig for energy-storage power stationsiPower signal distributionCoefficient, ΔP ci Representing energy storage power stationiAn active power difference control signal is generated in response to the active power difference control signal,sis a differential operator.
Establishing a state estimation model for the energy storage unit:
Figure 844658DEST_PATH_IMAGE038
（2）
in the formula (I), the compound is shown in the specification,SOC BESS (T+Δt) To representT+ΔtThe charge capacity of the SOC unit at the time,SOC BESS (T) RepresentTCharge capacity, Δ, of the time SOC celltFor a given time interval it is possible to provide,i o the current is output for the energy storage battery,ta time variable is represented by a time variable,C cap BESS_ the ampere capacity of the energy storage battery is obtained.
The constraints of the energy storage unit are as follows:
Figure 534001DEST_PATH_IMAGE039
（3）
in the formula (I), the compound is shown in the specification,SOC BESSmin the minimum charge capacity of the SOC is represented,SOC BESSmax represents the maximum charge capacity of the SOC,SOC BESSi representing energy storage power stationiThe SOC of (1) is set to be,P i_min indicating a minimum output active power limit and,P i_max the maximum output active power limit is indicated,P i representing energy storage power stationiThe SOC outputs active power.
Before participating in frequency modulation, the respective charge states of the energy storage units are detected according to the unitsSOCDeviation and output power limit adjustment output:
Figure 995069DEST_PATH_IMAGE040
（4）
in the formula,. DELTA.P i Representing energy storage power stationiActive power difference, Δ, of the SOC outputu BESSi For energy-storage power stationsiControl command, Δu sigSOC A charge state adjusting signal of the energy storage power station,
β BESS representing the response time constant of the energy storage plant.
In the step 2, the models of other devices of the power grid except the energy storage unit are established and respectively expressed as follows:
hydroelectric and thermal power unit speed regulator model:
Figure 97018DEST_PATH_IMAGE041
（5）
in the formula,. DELTA.K gi For hydroelectric and thermal power generating unitsiThe position variation of the speed regulator regulating valve,β gi_speed for hydroelectric and thermal power generating unitsiTime constant of governor, Δu Gi_sig For hydroelectric and thermal power generating unitsiThe control signal is sent to the computer system,α ci for hydroelectric and thermal power generating unitsiThe power signal distribution coefficient is set to be,L ci for hydroelectric and thermal power generating unitsiThe coefficient of the droop is,
Figure 10747DEST_PATH_IMAGE006
is a regioniHydroelectric and thermal power generator unitiThe frequency of (d); deltaP ci For hydroelectric and thermal power unitsiThe active power variation.
Hydroelectric and thermal engine turbine models:
Figure 489133DEST_PATH_IMAGE042
（6）
in the formula,. DELTA.P mwi For hydroelectric and thermal power unitsiThe amount of change in the output power,β gi_turb for hydroelectric and thermal power generating unitsiThe turbine time constant.
Secondary frequency modulation model of hydroelectric and thermal power machine:
Figure 70287DEST_PATH_IMAGE043
（7）
in the formula,. DELTA.P ACE_sig Secondary frequency modulation signals of the power grid are obtained; deltafIs the system frequency variation;K ci is the secondary frequency modulation coefficient.
The photovoltaic power station active power reduction is adopted to participate in system frequency adjustment, the photovoltaic power station is equivalent to a first-order inertia link, and a dynamic response model is as follows:
Figure 77557DEST_PATH_IMAGE044
（8）
in the formula,. DELTA.P iPV Representing photovoltaic power stationsiThe amount of power change, Δ, of the outputu i_sigPv Representing photovoltaic power stationsiThe control signal is sent to the computer system,α iPV for photovoltaic power stationsiPower signal division coefficient, ΔP ci Representing photovoltaic power stationsiAn active power difference control signal is generated,β iPV for photovoltaic power stationsiA response time constant.
The wind farm frequency response model may be expressed as:
Figure 744162DEST_PATH_IMAGE045
（9）
in the formula (I), the compound is shown in the specification,
Figure 757730DEST_PATH_IMAGE046
is a reference power for the wind farm,
Figure 927811DEST_PATH_IMAGE012
for the reference frequency of the wind farm,
Figure 371562DEST_PATH_IMAGE013
is the actual angular frequency of the wind farm,
Figure 525463DEST_PATH_IMAGE014
which is the wind speed,C p is the wind energy utilization coefficient and the speed ratio of the wind energy utilization coefficient to the wind turbine blade tip
Figure 345651DEST_PATH_IMAGE047
Pitch angle
Figure 635818DEST_PATH_IMAGE016
In connection with this, the present invention is,
Figure 516050DEST_PATH_IMAGE048
in order to be the density of the air,Ris the radius of the area swept by the blade,J t is the inertia coefficient of the fan rotor and the generator.
In the step 3, the multi-station cooperative frequency modulation state space model is:
Figure 891667DEST_PATH_IMAGE049
（10）
in the formula (I), the compound is shown in the specification,H sys in order to obtain the integral inertia of the regional power grid,H iSG generating set for water and fireiThe inertia of the rotor is reduced to zero,H jWF for wind-power unitsjThe inertia of the rotor is reduced to zero,H kPV for photovoltaic unitskThe inertia of the rotor is reduced to zero,H mBESS for energy-storage power stationsmThe inertia of (a);S sys 、S Gi 、S jWF 、S kPV 、S mBESS respectively a system, a water and electricity and a thermal power generating unitiWind turbine generator setjPhotovoltaic unitkEnergy storage power stationmThe distribution coefficient of the inertia of (a),n s ，n w ，n p ，n e the number of the water-fire generator sets, the wind turbine generators, the photovoltaic generators and the energy storage power stations in the region is respectively.
The frequency response model is:
Figure 252897DEST_PATH_IMAGE050
（11）
in the formula,. DELTA.f(k+1) Is composed ofkFrequency difference, Δ, between +1 time and targetf(k) Is composed ofkFrequency difference, delta, of time of day and targetP WF (k)、ΔP PV (k)、ΔP BESS (k)、ΔP D (k) Respectively a wind power plant, a photovoltaic power plant, an energy storage, water and electricity and a thermal power generating unitkThe active margin of the moment in time,T T is a constant of proportionality that is,D f is the damping coefficient of the region or regions,P LN in order to be the amount of power disturbance,Has a function of the frequency response of the system,f n is the nominal frequency.
The overall state space equation for a regional power system can be expressed as:
Figure 663150DEST_PATH_IMAGE051
（12）
in the formula (I), the compound is shown in the specification,x(t) Is at the same timetThe overall state vector of the time system is,Ais a matrix of the whole system and is,Bis an integral control matrix and is characterized by that,B ω in the form of an overall perturbation matrix,M(t) Is at the same timetThe time of day is the overall control vector of the system,ω(t) Is at the same timetAnd (5) disturbance vectors of the whole system at the moment.
In the step 4, aiming at the problems of insufficient inertia and poor frequency dynamic response performance provided by the hydro-thermal power generating unit and the thermal power generating unit, a virtual inertia control link is designed. The method can simulate the inertia constant and the damping characteristic of the synchronous generator, thereby improving the robustness and the sensitivity of the system. The virtual inertial control element can be represented by the following equation:
Figure 307758DEST_PATH_IMAGE052
（13）
in the formula (I), the compound is shown in the specification,H i andD i is the virtual rotor parameter that is,P ref (s) The system is a Laplace transform function referenced by the inversion unit, represents the system energy shortage obtained by temporary calculation, and the output of the system energy shortage is transmitted to the wind, light, water, fire and storage and other cooperative frequency modulation units in a signal form; deltaf(s) Is represented bySystem frequency variation amount deltafThe laplacian transform function of (a).
The overall transfer function can be expressed as:
Figure 311617DEST_PATH_IMAGE053
（14）
in the formula (I), the compound is shown in the specification,R i andK i respectively, a virtual droop constant and a virtual integral gain.
Based on the above transfer function, the frequency segment taking into account the virtual inertia can be designed as follows:
Figure 332663DEST_PATH_IMAGE054
（15）
in the formula (I), the compound is shown in the specification,
Figure 597422DEST_PATH_IMAGE055
is the frequency variation of the system, ΔP 1 Active power regulation, delta, for regional wind, light, water and fire storageP 2 For uncontrolled disturbances, Δ, inside and outside the regionPFor regional out-bound active power output variations,J sys is a virtual inertia of the system and,
Figure 819456DEST_PATH_IMAGE025
in order to be the nominal frequency of the system,D sys is the damping coefficient of the system.
In the step 5, on the basis of the virtual inertia link, a multi-station coordinated frequency modulation state space model considering the virtual inertia ring is established, and the closed-loop controller is designed by adopting an H infinity robust method in consideration of frequent disturbance of renewable energy sourcesM=KxTo ensure that the system is disturbed by the outsideiAnd the stability under the condition is realized, the influence of disturbance on a regulated output function of the system is reduced to the maximum extent, and the closed-loop stable control of the frequency can be realized. The corresponding closed loop system:
Figure 169666DEST_PATH_IMAGE026
（16）
in the formula (I), the compound is shown in the specification,Min order to be a matrix of the states of the controller,xinputting a state vector for a systemoutput(k) Is composed ofkThe output function of the time of day system,C 1 in order to output the matrix for the state,D 12 in order to perturb the output matrix,D 11 in order to control the output matrix,Kin order to achieve the gain,x(k) Is a systemkThe state variable that is input at a time is,i(k) Is composed ofkDisturbance variable of the time system.
The robust performance constraint inequality of the system is:
Figure 869768DEST_PATH_IMAGE056
（17）
wherein, the first and the second end of the pipe are connected with each other,Ain the form of a matrix of parameters,Gin the form of a matrix of parameters,WandXis a positive definite matrix to be solved, T represents the matrix transposition,
Figure 986104DEST_PATH_IMAGE057
a parameter matrix is designed for the controller,
Figure 379040DEST_PATH_IMAGE029
for the controller transfer function minimum rejection ratio, Iis a matrix of the units,Zthe output matrix is evaluated for the controller, C 2 representing a state parameter matrix;
positive definite matrix if there is optimal solutionWAndXthe system is gradually stabilized and the gain parameter is controlled to beK=X(W) -1 I.e. byu(t)=X(W) -1 xFor co-operating frequency modulationHAnd an infinity controller. The linear matrix inequality is used for solving the control parameters, so that the solving complexity is reduced, and the optimal control parameters are ensured.
As shown in fig. 3, in step 6, an inertia control coefficient adaptive adjustment method is established based on the deep neural network method.
By utilizing the neural network, the operation state can be adjusted according to different conditions and the virtual inertia system can be controlled in real time. The input data to the neural network are the regional frequency deviation and frequency rate of change, and the input is converted to a two-dimensional mapping and pooling layer in the convolutional layer. And finally, a complete connection layer containing the output control signal and the system inertia parameter deviation. The convolutional layer contains a plurality of convolution kernels, and each element corresponds to a weight coefficient and a deviation value. The size of the convolution kernel is determined by the convolution kernel, and the description of the features of the normalization input data extracted by the convolution layer is as follows:
Figure 950966DEST_PATH_IMAGE058
（18）
in the formula (I), the compound is shown in the specification,O n j, the output of the convolutional layer is shown,fin order to map the layers of the image,
Figure 720339DEST_PATH_IMAGE031
representing a selected input mapping set, and generating an output matrix with a determined size through a selected input with a certain size and convolution kernel operation, and adding a plurality of weights and offset parameters;U i n,-1 represents a parameterized spatial filter, the specific actions of which are automatically learned from the data during the training of the network,M n ij, which represents the input of the convolutional layer,b n j, in order to make the real number bias term,
Figure 694112DEST_PATH_IMAGE059
represents a cross product;
the function of the pooling layer is to select features and filter layers of information from the convolved output data, which is expressed as follows:
Figure 257948DEST_PATH_IMAGE033
（19）
in the formula (I), the compound is shown in the specification,Q n+1,j the water is output from the pool layer,gin order to pool the mapping, the mapping is performed,k n+1,j andr n+1,j respectively representing the multiplicative deviation and the additive deviation,
Figure 317171DEST_PATH_IMAGE034
representing a pooling function;
the filter layer is thus represented as:
Figure 624655DEST_PATH_IMAGE060
（20）
in the formula (I), the compound is shown in the specification,p、qrespectively representing the length and width of the filter layer; the pooling operation is to reduce the output of the upper layers, which makes the transformation of the predicted input data more robust,J i,j is the output of the filtering layer, and is the output of the filtering layer,S p,q in order to filter the parameters of the filter,b i+p ,j+q--11 is a real number bias term;
finally, the complete connection layer expands the characteristic diagram into vectors and the output value of each complete connection layer; passing to the output requires defining a penalty function to accomplish this classification; the fully connected layer performs non-linear combination, and this process is described as follows:
Figure 444145DEST_PATH_IMAGE061
（21）
in the formula (I), the compound is shown in the specification,Y n representing the output of the fully connected layer(s),fa non-linear function representing a fully connected layer,w n andp n representing the weight and deviation of the fully connected layer, respectively.
The robust cooperative frequency modulation method considering the energy storage charge state and the adaptive inertia level is characterized in that wind, light, water and fire storage cooperative control is taken as a core, the charge state of an energy storage unit is kept balanced, frequency closed-loop control is realized, and the problem of adaptive inertia control adjustment is solved.

Claims (7)

1. The robust cooperative frequency modulation method considering the energy storage charge state and the self-adaptive inertia level is characterized by comprising the following steps of:
step 1, establishing an active-frequency response model of an energy storage unit participating in power grid frequency modulation, and further establishing a relation model of the state of charge and the output power of the energy storage unit;
step 2, considering other facilities influencing frequency in the power grid, and establishing other equipment models of the power grid except the energy storage unit;
step 3, constructing a multi-station cooperative frequency modulation state space model containing the wind, light, water and fire power storage system;
step 4, designing a virtual inertia control link on the basis of a multi-field station cooperative frequency modulation state space model to realize the simulation of frequency closed-loop control inertia;
step 5, establishing a multi-station cooperative frequency modulation state space model considering the virtual inertia ring, considering the frequent disturbance of the renewable energy source, and adopting
Figure 989333DEST_PATH_IMAGE001
A closed-loop controller is designed by a robust method to realize closed-loop stable control of frequency;
and 6, establishing an inertia control coefficient self-adaptive adjusting method based on a deep neural network method, and realizing the improvement of the dynamic response performance of the frequency under the multi-working-condition multi-disturbance scene.
2. The robust cooperative frequency modulation method taking energy storage state of charge and adaptive inertia level into account of claim 1, wherein:
in the step 1, an active-frequency response model of the energy storage unit participating in the frequency modulation of the power grid is established as follows:
Figure 100509DEST_PATH_IMAGE002
（1）
in the formula,. DELTA.P BESS_i For energy-storage power stationsiOutput active power variation, Δh BESSi_sig For energy-storage power stationsiThe control signal is sent to the computer system,α BESSi_sig for energy-storage power stationsiPower signal division factor, ΔP ci Representing energy storage power stationiAn active power difference control signal is generated in response to the active power difference control signal,β iBESS representing energy storage power stationiThe response time constant is set to be,sis a differential operator;
establishing a state estimation model for the energy storage unit:
Figure 48873DEST_PATH_IMAGE003
（2）
in the formula (I), the compound is shown in the specification,SOC BESS (T+Δt) To representT+ΔtThe charge capacity of the SOC unit at the time,SOC BESS (T) To representTCharge capacity, Δ, of the time-of-day SOC celltFor a given time interval it is possible to provide,i o in order to output the current for the energy storage battery,C cap BESS_ ampere capacity for energy storage batteries;
the constraint conditions of the energy storage unit are as follows:
Figure 688933DEST_PATH_IMAGE004
（3）
in the formula (I), the compound is shown in the specification,SOC BESSmin represents the minimum charge capacity of the SOC,SOC BESSmax represents the maximum charge capacity of the SOC,SOC BESSi representing energy storage power stationiThe SOC charge capacity of (a) is,P i_min indicating a minimum output active power limit and,P i_max the maximum output active power limit is indicated,P i representing energy storage power stationiThe SOC outputs active power;
before participating in frequency modulation, the respective charge states of the energy storage units are detected according to the unitsSOCDeviation and output power limit adjustment output:
Figure 988327DEST_PATH_IMAGE005
（4）
in the formula,. DELTA.P i Indicating energy storage unitsiDifference of active power output, deltau BESSi For energy storage cellsiControl command, Δu sigSOC For storing energyA state-of-charge adjustment signal for the cell,β BESS representing the response time constant of the energy storage unit.
3. The robust coordinated frequency modulation method taking into account energy storage state of charge and adaptive inertia level of claim 2, wherein:
in step 2, establishing models of other devices of the power grid except the energy storage unit, which are respectively expressed as follows:
hydroelectric and thermal power unit speed regulator model:
Figure 106456DEST_PATH_IMAGE006
（5）
in the formula,. DELTA.K gi For hydroelectric and thermal power generating unitsiThe position variation of the speed regulator regulating valve,α ci for hydroelectric and thermal power unitsiPower signal division factor, ΔP ci For hydroelectric and thermal power generating unitsiThe amount of change in the active power is,L ci for hydroelectric and thermal power generating unitsiThe coefficient of the droop is,
Figure 174906DEST_PATH_IMAGE007
is a regioniHydroelectric and thermal power generator unitiFrequency of (a)u Gi_sig For hydroelectric and thermal power generating unitsiThe control signal is sent to the computer system,β gi_speed for hydroelectric and thermal power generating unitsiA governor time constant;
hydroelectric and thermal power turbine models:
Figure 519956DEST_PATH_IMAGE008
（6）
in the formula,. DELTA.P mwi For hydroelectric and thermal power generating unitsiThe amount of change in the output power,β gi_turb for hydroelectric and thermal power generating unitsiA turbine time constant;
secondary frequency modulation model of hydroelectric and thermal power machine:
Figure 182012DEST_PATH_IMAGE009
（7）
in the formula,. DELTA.P ACE_sig Secondary frequency modulation signals of the power grid;K ci is the coefficient of quadratic frequency modulation, ΔfIs the system frequency variation;
the photovoltaic power station is adopted to actively reduce and participate in system frequency adjustment, the photovoltaic power station is equivalent to a first-order inertia link, and a dynamic response model is as follows:
Figure 369411DEST_PATH_IMAGE010
（8）
in the formula,. DELTA.P iPV Representing photovoltaic power stationsiActive variation of output, Δu i_sigPv Representing photovoltaic power stationsiThe control signal is sent to the computer system,α iPV for photovoltaic power stationsiPower signal division coefficient, ΔP ci Representing photovoltaic power stationsiThe active power control signal is a signal that controls the active power,β iPV for photovoltaic power stationsiA response time constant;
the wind farm frequency response model may be expressed as:
Figure 292368DEST_PATH_IMAGE011
（9）
in the formula (I), the compound is shown in the specification,
Figure 398864DEST_PATH_IMAGE012
is the actual angular frequency of the wind farm,J t which is the coefficient of inertia of the fan rotor and the generator,
Figure 407271DEST_PATH_IMAGE013
is the density of the air, and is,Ris the radius of the area swept by the blade,
Figure 398361DEST_PATH_IMAGE014
for the reference frequency of the wind farm,C p in order to obtain the coefficient of utilization of wind energy,
Figure 175824DEST_PATH_IMAGE015
the speed ratio of the blade tip of the wind turbine,
Figure 325658DEST_PATH_IMAGE016
to be the pitch angle,
Figure 821362DEST_PATH_IMAGE017
which is the wind speed,
Figure 350563DEST_PATH_IMAGE018
is the wind farm reference power.
4. The robust coordinated frequency modulation method taking into account energy storage state of charge and adaptive inertia level of claim 3, wherein:
in step 3, the multi-field station collaborative frequency modulation state space model is as follows:
Figure 248112DEST_PATH_IMAGE019
（10）
in the formula (I), the compound is shown in the specification,H sys is the integral inertia of the regional power grid,H iSG generating set for water and fireiThe inertia of the rotor is reduced to zero,H jWF for wind-power generation unitsjThe inertia of the rotor is reduced to zero,H kPV for photovoltaic unitskThe inertia of the rotor is reduced to zero,H mBESS for energy-storage power stationsmThe inertia of (a);S sys 、S Gi 、S jWF 、S kPV 、S BESSm respectively a system, a hydroelectric power unit and a thermal power unitiWind turbine generator setjPhotovoltaic unitkEnergy storage power stationmThe distribution coefficient of the inertia of (a),n s ，n w ，n p ，n e the number of the water-fire generator sets, the wind turbine generators, the photovoltaic generators and the energy storage power stations in the region is respectively;
the frequency response model is:
Figure 837356DEST_PATH_IMAGE020
（11）
in the formula,. DELTA.f(k+1) Is composed ofkFrequency difference, Δ, between +1 time and targetf(k) Is composed ofkFrequency difference, delta, between time and targetP WF (k)、ΔP PV (k)、ΔP BESS (k)、ΔP D (k) Respectively a wind power plant, a photovoltaic power plant, an energy storage, water and electricity and a thermal power generating unitkThe active margin of the moment in time,T T is a constant of proportionality that is,D f is the damping coefficient of the region or regions,P LN in order to be the amount of power disturbance,Has a function of the frequency response of the system,f n is a rated frequency;
the overall state space equation for a regional power system is expressed as:
Figure 820356DEST_PATH_IMAGE021
（12）
in the formula (I), the compound is shown in the specification,x(t) Is composed oftThe overall state vector of the time system is,Ais a matrix of the whole system, and the system,Bis an integral control matrix and is provided with a plurality of control matrixes,B ω in the form of an overall perturbation matrix,M(t) Is composed oftThe time of day is the overall control vector of the system,ω(t) Is composed oftAnd (5) disturbance vectors of the whole system at the moment.
5. The robust cooperative frequency modulation method taking energy storage state of charge and adaptive inertia level into account of claim 4, wherein:
the virtual inertia control element is represented by the following equation:
Figure 887669DEST_PATH_IMAGE022
（13）
in the formula,. DELTA.P ref (s) Being reference to inverting unitssThe transformation function represents the energy shortage of the system obtained by current calculation, the output of the transformation function is transmitted to the coordinated frequency modulation units such as wind, light, water, fire storage and the like in a signal form,H i andD i is a virtual rotor parameter; delta off(s) Is the system frequency variation deltafThe laplace transform function of;
the overall transfer function is expressed as:
Figure 374145DEST_PATH_IMAGE023
（14）
in the formula (I), the compound is shown in the specification,R i andK i respectively, a virtual droop constant and a virtual integral gain;
based on the above transfer function, the frequency link design considering virtual inertia is as follows:
Figure 227018DEST_PATH_IMAGE024
（15）
in the formula (I), the compound is shown in the specification,
Figure 431735DEST_PATH_IMAGE025
is the frequency variation of the system, ΔP 1 Active power regulation, delta, for regional wind, light, water, fire and storageP 2 For uncontrolled disturbances, Δ, inside and outside the regionPFor regional out-bound active power output variations,J sys is the virtual inertia of the system and is,
Figure 302739DEST_PATH_IMAGE026
in order to be the nominal frequency of the system,D sys is the damping coefficient of the system.
6. The robust cooperative frequency modulation method taking into account energy storage state of charge and adaptive inertia level of claim 5, wherein:
step 5, on the basis of the virtual inertia link, establishing a multi-station cooperative frequency modulation state space model considering the virtual inertia ring, and considering the frequent disturbance of the renewable energy sources, adopting an H infinity robust method to design a closed-loop controllerM=KxTo ensure the system to be disturbed by the outsideiStability under the condition, and furthest reduce disturbance and to the influence of system by the transfer output function, realize the closed loop stability control of frequency, corresponding closed loop system:
Figure 909301DEST_PATH_IMAGE027
（16）
in the formula (I), the compound is shown in the specification,Min order to control the gain matrix of the controller,xthe state vector is input for the system and,output(k) Is composed ofkThe output function of the time of day system,C 1 in order to output the matrix for the state,D 12 in order to perturb the output matrix,D 11 in order to control the output matrix,Kin order to achieve the gain,x(k) Is a systemkThe state variable that is input at a time is,i(k) Is composed ofkDisturbance variables of the time system;
the robust performance constraint inequality of the system is:
Figure 105927DEST_PATH_IMAGE028
（17）
wherein the content of the first and second substances,Ain the form of a matrix of parameters,Gin the form of a matrix of parameters,WandXis a positive definite matrix to be solved, T represents the matrix transposition,
Figure 797939DEST_PATH_IMAGE029
a parameter matrix is designed for the controller,
Figure 472634DEST_PATH_IMAGE030
for the controller transfer function minimum rejection ratio,Iis a matrix of the units,Zis a controllerThe output matrix is evaluated and the output matrix is evaluated,C 2 representing a state parameter matrix;
positive definite matrix if optimal solution existsWAndXthe system is gradually stabilized and the gain parameter is controlled to beK=X(W) -1 I.e. byu(t)=X(W) -1 xFor co-frequency modulationHAnd the infinity controller solves the control parameters by using a linear matrix inequality, thereby reducing the solving complexity and ensuring the optimal control parameters.
7. The robust coordinated frequency modulation method taking into account energy storage state of charge and adaptive inertia level of claim 6, wherein:
in step 6, establishing an inertia control coefficient self-adaptive adjusting method based on a deep neural network method, specifically comprising the following steps:
the operation state is adjusted and the real-time control of the virtual inertia system is realized according to different conditions by utilizing a neural network; the input data of the neural network are regional frequency deviation and frequency change rate, and the input is converted into a two-dimensional mapping and pooling layer in the convolutional layer; finally, a complete connection layer comprises an output control signal and system inertia parameter deviation; the convolutional layer comprises a plurality of convolution kernels, each element corresponds to a weight coefficient and a deviation value, the size of each element is determined by the convolution kernels, and the features of the normalized input data extracted by the convolutional layer are described as follows:
Figure 199282DEST_PATH_IMAGE031
（18）
in the formula (I), the compound is shown in the specification,O n j, the output of the convolution layer is shown,fin order to map the layers of the image,
Figure 32721DEST_PATH_IMAGE032
representing a selected input mapping set, and generating an output matrix with a determined size through a selected input with a certain size and convolution kernel operation, and adding a plurality of weights and offset parameters;U i n,-1 representParameterizing a spatial filter, the specific action of which is automatically learned from the data during the training of the network,M n ij, which represents the input of the convolutional layer,b n j, in order to make the real number bias term,
Figure 477609DEST_PATH_IMAGE033
represents a cross product;
the function of the pooling layer is to select features and filter layers of information from the convolved output data, which is expressed as follows:
Figure 955995DEST_PATH_IMAGE034
（19）
in the formula (I), the compound is shown in the specification,Q n+1,j the water is output from the pond-forming layer,gin order to pool the mapping, the mapping is performed,k n+1,j andr n+1,j respectively representing the multiplicative deviation and the additive deviation,
Figure 271569DEST_PATH_IMAGE035
representing a pooling function;
the filter layer is thus represented as:
Figure 544419DEST_PATH_IMAGE036
（20）
in the formula (I), the compound is shown in the specification,p、qrespectively representing the length and width of the filter layer; the pooling operation is to reduce the output of the upper layers, which makes the transformation of the predicted input data more robust,J i,j is output for the filtering layer, and is output by the filtering layer,S p,q in order to filter the parameters of the process,b i+p ,j+q--11 is a real number bias term;
finally, a complete connection layer expands the characteristic diagram into a vector and an output value of each complete connection layer; passing to the output requires defining a loss function to accomplish this classification; the fully connected layer performs non-linear combination, and this process is described as follows:
Figure 945444DEST_PATH_IMAGE037
（21）
in the formula (I), the compound is shown in the specification,Y n represents the output of the fully-connected layer,fa non-linear function representing the fully connected layer,w n andp n representing the weight and deviation of the fully connected layer, respectively.
CN202211226501.2A 2022-10-09 2022-10-09 Robust cooperative frequency modulation method considering energy storage charge state and adaptive inertia level Active CN115296308B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202211226501.2A CN115296308B (en) 2022-10-09 2022-10-09 Robust cooperative frequency modulation method considering energy storage charge state and adaptive inertia level

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202211226501.2A CN115296308B (en) 2022-10-09 2022-10-09 Robust cooperative frequency modulation method considering energy storage charge state and adaptive inertia level

Publications (2)

Publication Number Publication Date
CN115296308A true CN115296308A (en) 2022-11-04
CN115296308B CN115296308B (en) 2023-03-14

Family

ID=83833339

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202211226501.2A Active CN115296308B (en) 2022-10-09 2022-10-09 Robust cooperative frequency modulation method considering energy storage charge state and adaptive inertia level

Country Status (1)

Country Link
CN (1) CN115296308B (en)

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115642616A (en) * 2022-12-26 2023-01-24 湖北工业大学 Multi-target LFC control method considering BESS participation degree grouping optimization
CN116054204A (en) * 2023-03-03 2023-05-02 上海交通大学 Ship variable-frequency load motor-capacitor inertia frequency modulation cooperative control method
CN116073439A (en) * 2023-04-06 2023-05-05 湖南大学 Distributed multi-source power generation system and method adopting synchronous motor interface
CN116111616A (en) * 2023-04-13 2023-05-12 清华大学 Multi-time space scale power system frequency full-track coordination optimization control method
CN116667389A (en) * 2023-07-25 2023-08-29 昆明理工大学 Novel wind-storage combined prediction correction control method for inertia lifting of power system
CN117543628A (en) * 2024-01-08 2024-02-09 国网江苏省电力有限公司镇江供电分公司 Photovoltaic power station partitioning method and system

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111864813A (en) * 2020-06-23 2020-10-30 国网辽宁省电力有限公司电力科学研究院 Wind/thermal power combined frequency control method based on virtual weight coefficient
CN112398166A (en) * 2020-11-09 2021-02-23 西安热工研究院有限公司 Parameter analysis method for energy storage primary frequency modulation virtual synchronous machine
CN112636398A (en) * 2020-12-23 2021-04-09 东南大学 Wind-fire-storage combined secondary frequency modulation method
CN113013896A (en) * 2021-03-08 2021-06-22 河海大学 Coordination control method for jointly participating in primary frequency modulation of power grid by light storage
WO2021164112A1 (en) * 2020-02-18 2021-08-26 山东大学 Frequency control method and system during using wind farm as black-start power source by means of optimal configuration of energy storage
CN113824156A (en) * 2021-10-26 2021-12-21 国网新源控股有限公司 Virtual inertia and virtual droop-based pumped storage and battery energy storage cooperative frequency modulation method

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2021164112A1 (en) * 2020-02-18 2021-08-26 山东大学 Frequency control method and system during using wind farm as black-start power source by means of optimal configuration of energy storage
CN111864813A (en) * 2020-06-23 2020-10-30 国网辽宁省电力有限公司电力科学研究院 Wind/thermal power combined frequency control method based on virtual weight coefficient
CN112398166A (en) * 2020-11-09 2021-02-23 西安热工研究院有限公司 Parameter analysis method for energy storage primary frequency modulation virtual synchronous machine
CN112636398A (en) * 2020-12-23 2021-04-09 东南大学 Wind-fire-storage combined secondary frequency modulation method
CN113013896A (en) * 2021-03-08 2021-06-22 河海大学 Coordination control method for jointly participating in primary frequency modulation of power grid by light storage
CN113824156A (en) * 2021-10-26 2021-12-21 国网新源控股有限公司 Virtual inertia and virtual droop-based pumped storage and battery energy storage cooperative frequency modulation method

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
李军徽等: "基于权重因子和荷电状态恢复的储能系统参与一次调频策略", 《电力系统自动化》 *
田蓓等: "可再生能源发电系统虚拟惯量动态控制仿真模型", 《可再生能源》 *

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115642616A (en) * 2022-12-26 2023-01-24 湖北工业大学 Multi-target LFC control method considering BESS participation degree grouping optimization
CN116054204A (en) * 2023-03-03 2023-05-02 上海交通大学 Ship variable-frequency load motor-capacitor inertia frequency modulation cooperative control method
CN116054204B (en) * 2023-03-03 2024-01-30 上海交通大学 Ship variable-frequency load motor-capacitor inertia frequency modulation cooperative control method
CN116073439A (en) * 2023-04-06 2023-05-05 湖南大学 Distributed multi-source power generation system and method adopting synchronous motor interface
CN116111616A (en) * 2023-04-13 2023-05-12 清华大学 Multi-time space scale power system frequency full-track coordination optimization control method
CN116667389A (en) * 2023-07-25 2023-08-29 昆明理工大学 Novel wind-storage combined prediction correction control method for inertia lifting of power system
CN116667389B (en) * 2023-07-25 2023-10-13 昆明理工大学 Novel wind-storage combined prediction correction control method for inertia lifting of power system
CN117543628A (en) * 2024-01-08 2024-02-09 国网江苏省电力有限公司镇江供电分公司 Photovoltaic power station partitioning method and system
CN117543628B (en) * 2024-01-08 2024-03-26 国网江苏省电力有限公司镇江供电分公司 Photovoltaic power station partitioning method and system

Also Published As

Publication number Publication date
CN115296308B (en) 2023-03-14

Similar Documents

Publication Publication Date Title
CN115296308B (en) Robust cooperative frequency modulation method considering energy storage charge state and adaptive inertia level
Khooban et al. A novel deep reinforcement learning controller based type-II fuzzy system: Frequency regulation in microgrids
Abazari et al. Coordination strategies of distributed energy resources including FESS, DEG, FC and WTG in load frequency control (LFC) scheme of hybrid isolated micro-grid
Li et al. A coordinated dispatch method with pumped-storage and battery-storage for compensating the variation of wind power
CN110581571A (en) dynamic optimization scheduling method for active power distribution network
CN109412158B (en) Energy-abandon cost constraint-considered combined operation control method for sending-end power grid unit
Taghizadeh et al. Frequency control of isolated WT/PV/SOFC/UC network with new control strategy for improving SOFC dynamic response
CN114865701A (en) Wind storage combined frequency modulation method based on adaptive model predictive control
Magdy et al. Effective Control of Smart Hybrid Power Systems: Cooperation of Robust LFC and Virtual Inertia Control Systems
CN112994051B (en) Seawater pumped storage power station capacity optimization configuration method containing variable speed unit
CN115149580A (en) Wind, light, water, fire and storage combined secondary frequency modulation method considering uncertainty delay
CN115296309B (en) Wind, light, water, fire and storage combined secondary frequency modulation method based on real-time inertia estimation
CN116526511B (en) Method for controlling load frequency of multi-source cooperative participation system
CN115663849B (en) Method and controller for cooperative and rapid regulation and control of hydraulic power plant and industrial temperature control load
Peng et al. Coordinated AGC control strategy for an interconnected multi-source power system based on distributed model predictive control algorithm
Zhang et al. Hierarchical MPC-based energy management and frequency regulation participation of a virtual power plant
Liu et al. Coordinated Frequency Regulation Strategy of Pumped Storage Units and Battery Energy Storage System
CN112600202B (en) Method for calculating optimal power flow of power grid with controllable phase shifter considering randomness of new energy
CN115001046A (en) Double-layer optimization control method for participating in peak shaving and frequency modulation of multi-energy-storage power station
CN115102228A (en) Multi-target coordination frequency optimization method and device for wind power plant containing flywheel energy storage
CN114938040A (en) Comprehensive optimization regulation and control method and device for source-network-load-storage alternating current and direct current system
Feng et al. Nonlinear model predictive control for pumped storage plants based on online sequential extreme learning machine with forgetting factor
Miracle et al. Hybrid Metaheuristic Model for Optimal Economic Load Dispatch in Renewable Hybrid Energy System
Rajan et al. Small-signal stability analysis and frequency regulation strategy for photovoltaic sources in interconnected power system
Feng et al. Advantage of battery energy storage systems for assisting hydropower units to suppress the frequency fluctuations caused by wind power variations

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