CN118100156A - On-line evaluation method for wind power plant frequency supporting capability based on Koopman operator - Google Patents
On-line evaluation method for wind power plant frequency supporting capability based on Koopman operator Download PDFInfo
- Publication number
- CN118100156A CN118100156A CN202410234523.6A CN202410234523A CN118100156A CN 118100156 A CN118100156 A CN 118100156A CN 202410234523 A CN202410234523 A CN 202410234523A CN 118100156 A CN118100156 A CN 118100156A
- Authority
- CN
- China
- Prior art keywords
- koopman
- power plant
- wind power
- wind
- fan
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000011156 evaluation Methods 0.000 title claims abstract description 70
- 238000000034 method Methods 0.000 claims abstract description 45
- 238000011158 quantitative evaluation Methods 0.000 claims abstract description 37
- 230000009466 transformation Effects 0.000 claims abstract description 11
- 230000006870 function Effects 0.000 claims description 60
- 230000004044 response Effects 0.000 claims description 23
- 238000004590 computer program Methods 0.000 claims description 10
- 230000001360 synchronised effect Effects 0.000 claims description 6
- 239000006185 dispersion Substances 0.000 claims description 3
- XOFYZVNMUHMLCC-ZPOLXVRWSA-N prednisone Chemical compound O=C1C=C[C@]2(C)[C@H]3C(=O)C[C@](C)([C@@](CC4)(O)C(=O)CO)[C@@H]4[C@@H]3CCC2=C1 XOFYZVNMUHMLCC-ZPOLXVRWSA-N 0.000 claims description 3
- 230000008569 process Effects 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 description 11
- 238000012549 training Methods 0.000 description 11
- 230000008859 change Effects 0.000 description 9
- 239000011159 matrix material Substances 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 230000035699 permeability Effects 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000011157 data evaluation Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000006116 polymerization reaction Methods 0.000 description 1
- 238000011002 quantification Methods 0.000 description 1
- 238000013139 quantization Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 238000007665 sagging Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Classifications
-
- 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
- H02J3/38—Arrangements for parallely feeding a single network by two or more generators, converters or transformers
- H02J3/381—Dispersed generators
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
- H02J2203/20—Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J2300/00—Systems for supplying or distributing electric power characterised by decentralized, dispersed, or local generation
- H02J2300/20—The dispersed energy generation being of renewable origin
- H02J2300/28—The renewable source being wind energy
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Power Engineering (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Public Health (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention discloses a method for online assessment of wind power plant frequency supporting capacity based on Koopman operator, which belongs to the technical field of wind power plant control and comprises the following steps: acquiring a Koopman dictionary function according to a power grid structure and a wind farm structure, and performing dimension lifting transformation and fitting on historical running state data of the wind farm to obtain a Koopman operator; based on the Koopman operator column, writing linear programming equation constraint, based on wind power plant safe operation condition column, writing inequality constraint, and based on the quantitative evaluation index column, writing objective function to construct a programming model; inputting the collected current running state data of the wind power plant as a time sequence initial value into the planning model for solving to obtain quantitative evaluation indexes of multiple time scales and multiple energy forms and multiple evaluation layers; and the wind power plant frequency supporting capacity is evaluated by using the quantitative evaluation index, so that the wind power plant frequency supporting capacity can be rapidly and accurately quantized, and each fan can distribute frequency modulation power according to energy when the frequency of the station is modulated.
Description
Technical Field
The invention belongs to the technical field of wind power plant control, and particularly relates to a method for online evaluation of wind power plant frequency supporting capacity based on Koopman operator.
Background
Frequency safety is one of the important issues of safety and stability of power systems. The permeability of new energy represented by wind power is rapidly increased, the inertia characteristics of the power grid are changed, and complexity is injected into the problem of frequency stability. Meanwhile, in order to keep the frequency of the novel power system stable, new energy stations such as wind power plants and the like are required to have certain frequency supporting capability. However, since the frequency supporting capability of the wind farm is affected by wind resources and the own running state, etc., it is difficult to quantitatively evaluate.
In recent years, the evaluation method of the frequency supporting capability of the wind power plant is mainly divided into two types of model driving and data driving. The model driving method is mainly used for calculating an expression of a correlation coefficient or performing complex nonlinear programming calculation through a simplified wind power plant frequency support model, so that a frequency support capacity evaluation index is obtained. However, a large number of wind farm model parameters are needed by a similar method, a certain error exists between a result obtained by the existing parameter estimation method and the real parameters, part of mechanisms are simplified by the model, the evaluation speed is reduced by complex nonlinear calculation during evaluation, and the requirement of online evaluation is difficult to meet. The data driving method adopts a neural network and other fitting modes to carry out fitting calculation on the historical frequency modulation data to obtain a data fitting model, thereby rapidly calculating the defined evaluation index. However, the existing data driving method is separated from a mathematical model, the internal relation of the system is difficult to reflect, the interpretability of the evaluation result is to be questionable, and in the actual operation of the wind farm, the frequency modulation data in the limit scene is difficult to collect, so that the accuracy of the training result in the global range is difficult to ensure.
Therefore, the prior research on the frequency supporting capability of the wind power plant does not provide a practical online evaluation method. In summary, the existing wind power plant frequency supporting capacity quantifying method is not accurate enough in evaluation result and not fast in evaluation speed, and is difficult to meet the requirement of online evaluation.
Disclosure of Invention
Aiming at the defects or improvement demands of the prior art, the invention provides an online evaluation method for the frequency supporting capability of a wind power plant based on a Koopman operator, which aims at adopting a Koopman dictionary function to fit the Koopman operator to historical running state data so as to construct a linear programming model, solving the linear programming model to obtain quantitative indexes of multiple time scales and multiple energy forms and multiple evaluation levels, and rapidly and accurately quantifying the size of the frequency supporting capability of the wind power plant, thereby solving the technical problems of low accuracy of evaluation results and low evaluation speed of the traditional wind power plant frequency supporting capability quantification method.
To achieve the above object, according to one aspect of the present invention, there is provided a method for online assessment of wind farm frequency support capability based on Koopman operator, comprising:
S1: acquiring a Koopman dictionary function according to a power grid structure and a wind farm structure;
S2: performing dimension lifting transformation on historical operation state data of a wind power plant by using the Koopman dictionary function to obtain a high-dimensional data set, and fitting the high-dimensional data set to obtain a Koopman operator;
s3: based on the Koopman operator column, writing linear programming equation constraint, based on wind power plant safe operation condition column, writing inequality constraint, and based on the quantitative evaluation index column, writing objective function to construct a programming model; the quantitative evaluation index is used for representing the frequency supporting potential and the actual maximum output of the fan in the station;
s4: collecting current running state data of the wind power plant;
S5: inputting the current running state data of the wind power plant as a time sequence initial value into the planning model for solving to obtain quantitative evaluation indexes of multiple time scales and multiple energy forms and multiple evaluation levels; and evaluating the frequency supporting capability of the wind power plant by utilizing the quantitative evaluation index.
In one embodiment, the S1 includes:
Using the formula The Koopman dictionary function g is written in columns (x t);xt is the original part of the dictionary function; ψ (x t) is the up-dimensional part of the dictionary function, expressed as:
Wherein V w is wind speed, ΔP L is system power disturbance, k dr is droop coefficient, ω r is fan speed, ΔP e is output power deviation, Δf is system frequency deviation, ΔP g is equivalent synchronous machine speed regulator power deviation, and T e is electromagnetic torque of the fan; the superscript T denotes a transpose; the subscript t represents the moment, and subscripts i and j represent serial numbers of different fans in the wind farm.
In one embodiment, the step S3 includes:
s31: writing linear programming equation constraint based on the Koopman operator column;
s32: based on wind farm safe operation conditions, the inequality constraint is written;
S33: using the formula Representing an objective function to construct a planning model;
Wherein delta E k rel is the maximum release of the actual rotor kinetic energy of the wind power plant, J is the mechanical inertia constant of a single fan, For the initial rotational speed of fan i,/>For the lowest rotating speed in the response process of the fan i,/>For maximum release of actual standby power of a wind farm, T del2 is a primary frequency modulation time scale, P m is fan mechanical power, P 0 is fan initial power, N t is the number of time sequences after T del2 difference dispersion,/>The square of the corresponding rotation speed of the fan i at the moment N t.
In one embodiment, the step S31 includes:
The linear programming equation constraints based on Koopman operator column writing include: variable initial equality constraints and variable non-initial equality constraints;
The variable non-initial equation constraint is expressed as: g (x t)＝Kt·g(x0); wherein g (x t) is a Koopman dictionary function corresponding to an original part x t of the dictionary function, K t is the Koopman dictionary function corresponding to an initial value x 0 of the dictionary function, and K 0 is the Koopman dictionary function corresponding to the initial value x 0 of the dictionary function.
In one embodiment, the S32 includes:
the inequality constraint based on wind farm safe operation condition column writing includes: rotational speed constraint omega rminset≤ωr≤ωrmaxset and load constraint And converter capacity constraint-P rmaxset≤Pr≤Prmaxset;
Wherein omega r is the fan speed, omega rminset is the lower limit of the speed, omega rmaxset is the upper limit of the speed, T e is the load, P e is the output power, T emaxset is the upper limit of the load, P r is the converter capacity, P rmaxset is a set upper limit for the converter capacity.
In one embodiment, the quantitative evaluation index of the multi-time scale multi-energy form multi-evaluation level in S5 includes: the potential value of the primary time scale standby power average value and the potential utilization rate thereof, the potential value of the standby power response speed and the potential utilization rate of the standby power response speed.
In one embodiment, the potential utilization is a ratio of actual values to potential values.
According to another aspect of the present invention, there is provided an online evaluation device for supporting capability of wind power plant frequency based on Koopman operator, including:
The acquisition module is used for acquiring a Koopman dictionary function according to the power grid structure and the wind farm structure;
The fitting module is used for carrying out dimension lifting transformation on the historical running state data of the wind power plant by utilizing the Koopman dictionary function to obtain a high-dimensional data set, and fitting the high-dimensional data set to obtain a Koopman operator;
the modeling module is used for writing linear programming equation constraint based on the Koopman operator column, writing inequality constraint based on the wind power plant safe operation condition column, and writing objective function based on the quantitative evaluation index column so as to construct a programming model; the quantitative evaluation index is used for representing the frequency supporting potential and the actual maximum output of the fan in the station;
the acquisition module is used for acquiring current running state data of the wind power plant;
The evaluation module is used for inputting the current running state data of the wind power plant as a time sequence initial value into the planning model for solving to obtain quantitative evaluation indexes of multiple evaluation levels in a multi-time scale and multi-energy form; and evaluating the frequency supporting capability of the wind power plant by utilizing the quantitative evaluation index.
According to another aspect of the invention there is provided a wind farm control system comprising a memory storing a computer program and a processor implementing the steps of the above method when the processor executes the computer program.
According to another aspect of the present invention there is provided a computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of the above method.
In general, the above technical solutions conceived by the present invention, compared with the prior art, enable the following beneficial effects to be obtained:
(1) The invention provides an online evaluation method for the frequency supporting capability of a wind power plant based on a Koopman operator, which is characterized in that a Koopman dimension-increasing function is adopted to fit historical running state data, complex nonlinear time sequence data are converted into a high-dimensional linear time sequence data set to obtain a corresponding Koopman operator, a linear model for evaluating the frequency supporting capability of the wind power plant is constructed based on the Koopman operator, and the linear model is solved to obtain quantitative evaluation indexes of multiple time scales and multiple energy forms and multiple evaluation layers, so that the frequency supporting capability of the wind power plant can be rapidly and accurately quantized. According to the method, the frequency active frequency supporting capacity of the wind power plant can be rapidly and accurately calculated on line, frequency modulation power can be distributed to each fan when the frequency is modulated at the site, frequency modulation resources can be reasonably distributed to the power grid according to the evaluation result, waste of the resources is avoided, frequency safety of a power system is further improved, and engineering practicability is high.
(2) The scheme provides an online evaluation method for the frequency supporting capability of a wind power plant based on Koopman operator, which utilizes a formulaAnd (3) writing the Koopman dictionary function g (x t), fully considering the structures of the wind power plant and the power grid, enabling the Koopman operator obtained by fitting to be more accurate, and further improving the evaluation accuracy of the frequency supporting capability of the wind power plant.
(3) The scheme provides an online evaluation method for the frequency supporting capability of a wind power plant based on Koopman operator, which utilizes a formulaRepresenting an objective function to construct a planning model; the actual operation mechanism of the wind power plant is fully considered by the objective function corresponding to the planning model, so that the pertinence, the credibility and the accuracy of analysis are improved. The method has the advantages that the method has extremely high effect in the evaluation of the frequency supporting capability of the wind power plant, and the high reliability and practicability of the evaluation result are ensured.
(4) The scheme provides an online evaluation method for the frequency supporting capability of a wind power plant based on a Koopman operator, which utilizes the non-initial equation constraint of a Koopman description variable to characterize the linear programming equation constraint, so that the system is linearized in a high dimension, and the planning calculation efficiency is improved.
(5) The scheme provides an online evaluation method for the frequency supporting capability of a wind power plant based on Koopman operator, wherein the inequality constraint based on safe operation condition column writing of the wind power plant comprises the following steps: the rotation speed constraint, the load constraint and the converter capacity constraint not only consider the capacities of the fans in different running states, but also consider the limitation of safe running constraint, and the omnibearing consideration ensures that the evaluation result is more accurate and reliable.
(6) The scheme provides an online evaluation method for the frequency supporting capability of a wind power plant based on Koopman operator, wherein quantitative evaluation indexes comprise: the potential value of the primary time scale standby power average value, the potential utilization rate of the primary time scale standby power average value, the potential value of the standby power response speed and the potential utilization rate of the standby power response speed more comprehensively describe the frequency supporting capability of the wind power plant, so that the evaluation level is richer.
Drawings
FIG. 1 is a flowchart of a method for online evaluation of wind farm frequency support capability based on Koopman operator provided in embodiment 1 of the present invention;
fig. 2a and fig. 2b are schematic diagrams of a four-machine two-area system for wind farm access provided by an embodiment of the present invention;
FIG. 3a is a flowchart of a method for online evaluation of the frequency support capability of a wind farm based on Koopman operator according to embodiment 7 of the present invention;
FIG. 3b is a graph showing the variation of the kinetic energy release of the rotor with the wind speed according to embodiment 7 of the present invention;
FIG. 4 is a graph showing the response speed of the standby power according to embodiment 7 of the present invention;
FIG. 5 is a graph showing the average release of standby power as a function of wind speed according to example 7 of the present invention;
Fig. 6 is a graph of evaluation error provided in example 7 of the present invention.
Fig. 7 is a graph showing the variation of the safe operation constraint amount provided in embodiment 7 of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present invention more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention. In addition, the technical features of the embodiments of the present invention described below may be combined with each other as long as they do not collide with each other.
Example 1
In this embodiment, as shown in fig. 1, there is provided an online evaluation method for wind farm frequency supporting capability based on Koopman operator, including: s1: obtaining Koopman dictionary functions from grid structure and wind farm structure, it should be noted that grid structure and wind farm structure changes include, but are not limited to: fan switching, synchronous machine switching, change of wind farm load shedding coefficient and the like. S2: and carrying out dimension lifting transformation on the historical running state data of the wind power plant by using the Koopman dictionary function to obtain a high-dimension data set, and fitting the high-dimension data set to obtain the Koopman operator. S3: based on the Koopman operator column, linear programming equation constraint is written, based on wind farm safe operation condition column, inequality constraint is written, and based on the quantitative evaluation index column, an objective function is written, so that a programming model is constructed. The quantitative evaluation index is used for representing the frequency supporting potential and the actual maximum output of the fan in the station. S4: collecting current operational state data of a wind farm, comprising: fan wind speed V w, system power disturbance Δp L, fan speed ω r, etc. S5: and (3) taking the current running state data of the wind power plant as a time sequence initial value, inputting the time sequence initial value into a planning model, and solving to obtain quantitative evaluation indexes of multiple time scales, multiple energy forms and multiple evaluation layers. And evaluating the frequency supporting capability of the wind power plant by using the quantitative evaluation index. The evaluation result is reported to the station and the power grid, the frequency supporting potential index can be used as a basis for reasonably distributing the supporting power of the fan in the station, and the actual index can be used as a reference for the station to output at the limit on the premise of safe operation.
Wherein the required historical operating state data comprises: under the control of different wind speeds V w, different system power disturbance delta P L and different droop coefficients k dr, the rotating speed omega r time sequence response, the output power deviation delta P e time sequence response, the system frequency deviation delta f time sequence response and the equivalent synchronous machine speed regulator power deviation delta P g time sequence response of each fan of the wind power plant. Besides, the electromagnetic torque T e time sequence response of each fan of the wind power plant is also needed, and the electromagnetic torque T e time sequence response can be directly calculated by the measured value
Example 2
On the basis of embodiment 1, S1 in this embodiment operates as follows: using the formulaColumn-written Koopman dictionary function g (x t).xt is the original part of the dictionary function. ψ (x t) is the up-dimensional part of the dictionary function, expressed as:
Wherein V w is wind speed, ΔP L is system power disturbance, k dr is droop coefficient, ω r is fan speed, ΔP e is output power deviation, Δf is system frequency deviation, ΔP g is equivalent synchronous machine speed regulator power deviation, and T e is electromagnetic torque of the fan. The superscript T denotes a transpose. The subscript t represents the moment, and subscripts i and j represent serial numbers of different fans in the wind farm.
Wherein the control variable v t and the state variable u t together comprise the training fit dataset x t＝[ut,vt]T:
The state variables evolve forward in time sequence under the action of the control variables, and each control variable is approximately unchanged in single frequency modulation response. Because of the global linearization of Koopman, the training fit dataset does not have to contain data in extreme operating scenarios.
Further, the results after the data up-conversion at time t and time t+1 are as follows:
Wherein, Q is the number of variables in the high dimensional space. Performing least square fitting on the dimension-rising sample to obtain/>The approximate fitting matrix K of the operator:
Wherein, Represents the Moore-Penrose inverse matrix of X 0lift, and/>
Example 3
On the basis of embodiment 1, S3 in this embodiment operates as follows: s31: linear programming equation constraints are written based on Koopman operator columns. S32: the inequality constraint is written based on wind farm safe operation conditions. S33: using the formulaRepresenting the objective function to construct a planning model. Wherein delta E krel is the maximum release of the actual rotor kinetic energy of the wind farm, J is the mechanical inertia constant of a single fan, and is/For the initial rotational speed of fan i,/>For the lowest rotating speed in the response process of the fan i,/>For maximum release of actual standby power of a wind farm, T del2 is a primary frequency modulation time scale, P m is fan mechanical power, P 0 is fan initial power, N t is the number of time sequences after T del2 difference dispersion,/>The square of the corresponding rotation speed of the fan i at the moment N t.
Example 4
On the basis of embodiment 3, S31 in this embodiment operates as follows: the linear programming equation constraints based on Koopman operator column writing include: variable initial equation constraints and variable non-initial equation constraints. The variable non-initial equation constraint is expressed as: g (x t)＝Kt·g(x0). Wherein g (x t) is a Koopman dictionary function corresponding to an original part x t of the dictionary function, K t is the Koopman dictionary function corresponding to an initial value x 0 of the dictionary function, and K 0 is the Koopman dictionary function corresponding to the initial value x 0 of the dictionary function.
Wherein, the variable initial constraint in the linear programming model is provided by the measured operation data:
The droop coefficient is regarded as a variable to be solved, and the initial value is not set. And then carrying out up-dimensional transformation on the initial value, wherein the up-dimensional transformation is still linear operation after the up-dimensional transformation because the elements in the initial value of the variable are mostly known.
Further, the equality constraint of the variables in the linear programming model is: and g (x t)＝Kt·g(x0), namely time sequence data at the time t can be obtained by calculating after linear transformation of the initial time rising dimension data according to the Koopman operator K.
Example 5
On the basis of embodiment 3, S32 in this embodiment operates as follows: the inequality constraint based on wind farm safe operation condition column writing includes: rotational speed constraint omega rminset≤ωr≤ωrmaxset and load constraintAnd converter capacity constraint-P rmaxset≤Pr≤Prmaxset; wherein omega r is the fan speed, omega rminset is the lower limit of the speed, omega rmaxset is the upper limit of the speed, T e is the load, P e is the output power, T emaxset is the upper limit of the load, P r is the converter capacity,P rmaxset is a set upper limit for the converter capacity.
Example 6
Based on embodiment 1, the quantitative evaluation indexes of the multi-time-scale multi-energy form multi-evaluation level in S5 in this embodiment include: the potential value of the primary time scale standby power average value and the potential utilization rate thereof, the potential value of the standby power response speed and the potential utilization rate of the standby power response speed.
The quantitative evaluation index comprises a maximum potential value of the releasable kinetic energy of the rotor under an inertial time scale: Omega rminset is the set safe lower rotational speed limit, typically taken as 0.7p.u.
Wherein the quantization evaluation index comprises a reserve power limit slope value:
Where C 2＝-0.0101,c1＝-0.0042β+0.2077,c0 = 0.0243 β -0.6140 is a linear expression for β after expansion of C P, T emaxset is a set torque upper limit, typically taken as 1.2p.u..
The quantitative evaluation index comprises a standby power release maximum potential value under a primary time scale:
Where P mppt is the maximum power output, P m0 is the derated power output, α is the correction factor, and k PFR is the limit response slope.
In addition, the quantitative evaluation index comprises the actual value calculation method of the standby power slope of the wind power plant as the actual measured power difference derivative. The method also comprises the maximum release of the actual rotor kinetic energy of the wind power plant and the maximum release of the actual standby power of the wind power plant.
Example 7
Based on embodiment 6, the potential utilization in this embodiment is the ratio of the actual value to the potential value.
The quantitative evaluation index of the multi-time-scale multi-energy form multi-evaluation level comprises potential utilization rate of each component and is calculated by the ratio of the actual value index to the potential value index.
A simulation experiment procedure is given below:
as shown in FIG. 3a, the method for online assessment of the wind farm frequency support capability based on the Koopman operator comprises the following steps.
S1: according to the power grid structure and the wind farm structure, a Koopman dictionary function is obtained, and the function structure is as follows: the data up-scaling transform is run for multi-state variable timing. ψ (x t) is the upbound part of the dictionary function:
s2: and fitting the historical frequency modulation data to obtain a Koopman operator of the system which is linearly evolved in a high-dimensional space. Further, the least square fitting is carried out on the dimension-rising sample to obtain The approximate fitting matrix K of the operator: /(I)Wherein, historical frequency modulation data is generated by simulation. The parameter setting ranges are as follows: the wind speed V w is set to be changed from 9.4m/s to 11.4m/s, the system power disturbance delta P L is 400MW, and the fan rotating speed omega r is changed along the wind speed range from 0.65p.u. to 0.9p.u. at the moment of frequency modulation according to the load shedding curve operation and overspeed load shedding point. Specifically, the control variable v t and the state variable u t together comprise a training fit dataset; x t＝[u,tv]T t: /(I)
S3: based on the Koopman operator column writing linear programming equation constraint, based on the wind power plant safe operation condition column writing inequality constraint, based on the quantitative evaluation index column writing objective function, the current operation state data is used as a time sequence initial value to be input into a programming model, and the construction of the linear programming model is completed. Specifically, the rotation speed safety lower limit omega rminset is set to be 0.7p.u., the load safety upper limit T emaxset is set to be 1.2p.u., the active available capacity P rmaxset of the converter is set to be 0.3p.u., and the maximum value of the objective function is calculated by substituting the maximum value into a linear programming solving model.
S4: collecting current operational state data of a wind farm, comprising: fan wind speed V w, system power disturbance Δp L, fan speed ω r, etc. The wind power plant comprises 250 doubly-fed machine sets with the capacity of 2MW, wherein the internal parameters of each machine set are the same, and a frequency modulation scheme combining overspeed load shedding and sagging control is adopted. To simplify the calculation, the whole wind farm is equivalent to a polymerization fan with the capacity of 500MW, and the wind farm and the system model for testing are shown in fig. 2a and 2 b. And changing the wind speed and the disturbance power to obtain each frequency modulation data.
And S5, solving a planning model, calculating quantitative evaluation indexes of multiple time scales and multiple energy forms and multiple evaluation layers, and reporting the power grid after finishing the result.
The maximum potential value calculation method of the releasable kinetic energy of the rotor under the inertia time scale comprises the following steps:
substituting the formula can calculate the maximum potential value of the releasable kinetic energy of the rotor under the inertia time scale, and the evaluation result of the index along with the change of the wind speed is shown in fig. 3 b.
The method for calculating the actual maximum release value of the releasable kinetic energy of the rotor under the inertia time scale comprises the following steps of: The linear programming calculates the evaluation of the index as a function of wind speed as shown in fig. 3 b.
The method for calculating the reserve power limit slope value comprises the following steps:
the evaluation result of the index along with the change of wind speed and rotation speed is shown in fig. 4.
The method for calculating the maximum potential value of the standby power release under the primary time scale comprises the following steps: the evaluation result of the index with the change of wind speed is shown in fig. 5.
The method for calculating the actual maximum release value of the standby power under the primary time scale comprises the following steps of:
the evaluation result of the linear programming calculation of the index with the change of wind speed is shown in fig. 5.
Specifically, for a determined operating state, the final reported evaluation index calculation result is shown in table 1:
table 1 frequency support capability quantitative assessment index for wind farms
As shown in fig. 2a and fig. 2b, the test system is a double-fed wind power plant connected into a four-machine two-area system, the synchronous generator adopts a three-step model, the double-fed wind power plant adopts a plurality of single DFIG aggregation equivalent values with rated power of 2MW, and the load adopts a constant impedance model. The bus 8 is connected into a doubly-fed wind power plant with rated power of 500MW, the load power is 1700MW, and the wind power permeability is 29.41%.
As shown in fig. 3b, as the wind speed increases, the potential and actual capacity of the wind farm inertia time scale rotor kinetic energy release gradually increases, on the one hand, the wind speed increases to increase the rotational speed corresponding to the load shedding curve, resulting in an increase in the available kinetic energy of the rotor, and on the other hand, the actual maximum capacity is lower than the maximum potential in a high wind speed range due to the influence of load, converter capacity.
As shown in fig. 4, as the wind speed and the rotation speed change, there is a negative value of the maximum change rate, because the working point of the fan deviates from the overspeed load shedding area, that is, moves to the left of the MPPT point, and the standby power stored in the fan is exhausted, which can be regarded as the capability of not having the frequency support of the primary time scale.
As shown in fig. 5, as the wind speed increases, the average value of the standby power of the wind farm in one time scale gradually increases, on one hand, the wind speed increases to increase the maximum output power corresponding to the MPPT point of the wind turbine, and the standby power is increased along with the increase of the maximum output power according to the load reduction, on the other hand, the standby power cannot always change according to the maximum response speed due to the limitation of the load constraint, so that the actual maximum capacity is always lower than the maximum potential.
The reliability of the method provided by the invention is further verified, and under consideration of different sample training ranges, the evaluation error calculation result of the Koopman high-dimensional linear programming method is compared. The wind speed range of the sample training set 1 is set to be 10.1 m/s-10.7 m/s, the wind speed range of the sample training set 2 is set to be 9.5 m/s-11.4 m/s, the ranges of the rest variables are the same, and the evaluation error is shown in figure 6. After the data of part of working conditions are subjected to Koopman global linearization, the working conditions of the global range can be estimated, so that data acquisition under a limiting working condition scene is avoided, and the acquisition difficulty of evaluation data is reduced. The range of the training sample training set is properly enlarged, so that the evaluation error can be effectively reduced, two factors of precision and data acquisition are required to be comprehensively considered, and the range of the sample training set is reasonably selected by combining the running condition of an actual wind field while avoiding the occurrence of the condition of the maximum error.
Further verifying the reliability of the method provided by the invention, taking a specific wind speed working condition of 10.7m/s, and the rotor rotating speed, the torque and the active response curve of the converter are shown in figure 7. At this time, the rotation speed constraint is dominant constraint, the rotation speed of the rotor reaches the lower limit of 0.7p.u., and the electromagnetic torque and the active power of the converter are still a certain margin from the boundary. Therefore, the boundary running state of the wind power plant can be accurately determined by the assessment method, and the frequency support capability index obtained based on the boundary running state shows higher reliability.
The invention discloses an online evaluation method for the frequency supporting capability of a wind power plant based on Koopman operator and multiple evaluation level indexes, which comprises the following steps: the method comprises the steps of quantifying inertial support capacity by rotor kinetic energy of an inertial scale, quantifying primary frequency modulation capacity by standby power of a primary scale, quantifying frequency modulation response speed by power limit change rate, and quantifying available capacity and real capacity respectively in potential and actual two layers. Furthermore, according to the data evaluation method based on the Koopman operator, the global linearization model of the evaluation object is obtained through training the historical frequency modulation data, so that the evaluation result is calculated in an online mode after the state data are synchronously measured. The test of the example system shows that the provided evaluation method can accurately quantify the frequency supporting capability of the wind power plant, has the characteristics of novelty, practicability, accuracy and high calculation efficiency, provides a reference for frequency modulation distribution scheduling, reasonably utilizes frequency modulation resources and improves the safety and stability of system frequency.
Example 8
In this embodiment, there is provided an apparatus for online assessment of wind farm frequency supporting capability based on Koopman operator, for executing the method for online assessment of wind farm frequency supporting capability in each of the above embodiments, where the apparatus includes: the system comprises an acquisition module, a fitting module, a modeling module, an acquisition module and an evaluation module. And the acquisition module is used for acquiring the Koopman dictionary function according to the power grid structure and the wind farm structure. And the fitting module is used for carrying out dimension lifting transformation on the historical running state data of the wind power plant by utilizing the Koopman dictionary function to obtain a high-dimension data set, and fitting the high-dimension data set to obtain the Koopman operator. The modeling module is used for writing linear programming equation constraint based on the Koopman operator column, writing inequality constraint based on the wind power plant safe operation condition column, and writing objective function based on the quantitative evaluation index column to construct a programming model. The quantitative evaluation index is used for representing the frequency supporting potential and the actual maximum output of the fan in the station. And the acquisition module is used for acquiring current running state data of the wind power plant. The evaluation module is used for inputting the current running state data of the wind power plant as a time sequence initial value into the planning model to solve, so as to obtain quantitative evaluation indexes of multiple time scales and multiple energy forms and multiple evaluation layers. And evaluating the frequency supporting capability of the wind power plant by using the quantitative evaluation index.
Example 9
In this embodiment, a wind farm control system is provided, including a memory and a processor, where the memory stores a computer program, and the processor implements the steps of the method when executing the computer program.
Example 10
In this embodiment, a computer-readable storage medium is provided, on which a computer program is stored which, when executed by a processor, implements the steps of the above-described method.
It will be readily appreciated by those skilled in the art that the foregoing description is merely a preferred embodiment of the invention and is not intended to limit the invention, but any modifications, equivalents, improvements or alternatives falling within the spirit and principles of the invention are intended to be included within the scope of the invention.
Claims (10)
1. The method for online evaluation of the wind power plant frequency supporting capability based on the Koopman operator is characterized by comprising the following steps of:
S1: acquiring a Koopman dictionary function of the Coumann according to the power grid structure and the wind farm structure;
S2: performing dimension lifting transformation on historical operation state data of a wind power plant by using the Koopman dictionary function to obtain a high-dimensional data set, and fitting the high-dimensional data set to obtain a Koopman operator;
s3: based on the Koopman operator column, writing linear programming equation constraint, based on wind power plant safe operation condition column, writing inequality constraint, and based on the quantitative evaluation index column, writing objective function to construct a programming model; the quantitative evaluation index is used for representing the frequency supporting potential and the actual maximum output of the fan in the station;
s4: collecting current running state data of the wind power plant;
S5: inputting the current running state data of the wind power plant as a time sequence initial value into the planning model for solving to obtain quantitative evaluation indexes of multiple time scales and multiple energy forms and multiple evaluation levels; and evaluating the frequency supporting capability of the wind power plant by utilizing the quantitative evaluation index.
2. The Koopman operator based wind farm frequency support capability online assessment method according to claim 1, wherein S1 comprises:
Using the formula The Koopman dictionary function g is written in columns (x t);xt is the original part of the dictionary function; ψ (x t) is the up-dimensional part of the dictionary function, expressed as:
Wherein V w is wind speed, ΔP L is system power disturbance, k dr is droop coefficient, ω r is fan speed, ΔP e is output power deviation, Δf is system frequency deviation, ΔP g is equivalent synchronous machine speed regulator power deviation, and T e is electromagnetic torque of the fan; the superscript T denotes a transpose; the subscript t represents the moment, and subscripts i and j represent serial numbers of different fans in the wind farm.
3. The Koopman operator based wind farm frequency support capability online assessment method according to claim 1, wherein S3 comprises:
s31: writing linear programming equation constraint based on the Koopman operator column;
s32: based on wind farm safe operation conditions, the inequality constraint is written;
S33: using the formula Representing an objective function to construct a planning model;
Wherein delta E k rel is the maximum release of the actual rotor kinetic energy of the wind power plant, J is the mechanical inertia constant of a single fan, For the initial rotational speed of fan i,/>For the lowest rotating speed in the response process of the fan i,/>For maximum release of actual standby power of a wind farm, T del 2 is a primary frequency modulation time scale, P m is fan mechanical power, P 0 is fan initial power, N t is the number of time sequences after T del 2 difference dispersion,/>The square of the corresponding rotation speed of the fan i at the moment N t.
4. The Koopman operator based wind farm frequency support capability online assessment method according to claim 3, wherein S31 comprises:
The linear programming equation constraints based on Koopman operator column writing include: variable initial equality constraints and variable non-initial equality constraints;
The variable non-initial equation constraint is expressed as: g (x t)＝Kt·g(x0); wherein g (x t) is a Koopman dictionary function corresponding to an original part x t of the dictionary function, K t is the Koopman dictionary function corresponding to an initial value x 0 of the dictionary function, and K 0 is the Koopman dictionary function corresponding to the initial value x 0 of the dictionary function.
5. The Koopman operator based wind farm frequency support capability online assessment method according to claim 3, wherein S32 comprises:
the inequality constraint based on wind farm safe operation condition column writing includes: rotational speed constraint omega rminset≤ωr≤ωrmaxset and load constraint And converter capacity constraint-P r max set≤Pr≤Pr max set;
Wherein omega r is the fan speed, omega rminset is the lower limit of the speed, omega rmaxset is the upper limit of the speed, T e is the load, P e is the output power, T emaxset is the upper limit of the load, P r is the converter capacity, P r max set is a set upper limit for the converter capacity.
6. The Koopman operator based wind farm frequency support capability online assessment method according to claim 1, wherein the quantitative assessment index of the multi-time scale multi-energy form multi-assessment hierarchy in S5 comprises: the potential value of the primary time scale standby power average value and the potential utilization rate thereof, the potential value of the standby power response speed and the potential utilization rate of the standby power response speed.
7. The Koopman operator based wind farm frequency support capability online assessment method according to claim 6, wherein the potential utilization is a ratio of an actual value to a potential value.
8. An online evaluation device for wind power plant frequency supporting capability based on Koopman operator, which is characterized by comprising:
The acquisition module is used for acquiring a Koopman dictionary function of the Coulomb according to the power grid structure and the wind power plant structure;
The fitting module is used for carrying out dimension lifting transformation on the historical running state data of the wind power plant by utilizing the Koopman dictionary function to obtain a high-dimensional data set, and fitting the high-dimensional data set to obtain a Koopman operator;
the modeling module is used for writing linear programming equation constraint based on the Koopman operator column, writing inequality constraint based on the wind power plant safe operation condition column, and writing objective function based on the quantitative evaluation index column so as to construct a programming model; the quantitative evaluation index is used for representing the frequency supporting potential and the actual maximum output of the fan in the station;
the acquisition module is used for acquiring current running state data of the wind power plant;
The evaluation module is used for inputting the current running state data of the wind power plant as a time sequence initial value into the planning model for solving to obtain quantitative evaluation indexes of multiple evaluation levels in a multi-time scale and multi-energy form; and evaluating the frequency supporting capability of the wind power plant by utilizing the quantitative evaluation index.
9. A wind farm control system comprising a memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, carries out the steps of the method according to any of claims 1 to 7.
10. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410234523.6A CN118100156A (en) | 2024-02-29 | 2024-02-29 | On-line evaluation method for wind power plant frequency supporting capability based on Koopman operator |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410234523.6A CN118100156A (en) | 2024-02-29 | 2024-02-29 | On-line evaluation method for wind power plant frequency supporting capability based on Koopman operator |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN118100156A true CN118100156A (en) | 2024-05-28 |
Family
ID=91152881
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410234523.6A Pending CN118100156A (en) | 2024-02-29 | 2024-02-29 | On-line evaluation method for wind power plant frequency supporting capability based on Koopman operator |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118100156A (en) |
-
2024
- 2024-02-29 CN CN202410234523.6A patent/CN118100156A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109038613A (en) | A kind of adaptive low frequency deloading method counted and wind-powered electricity generation virtual inertia/primary frequency modulation responds | |
| CN110350518B (en) | Power grid energy storage capacity demand assessment method and system for peak shaving | |
| CN107947228B (en) | Stochastic stability analysis method for power system containing wind power based on Markov theory | |
| CN110061521B (en) | Maximum wind power permeability rapid evaluation method considering frequency accumulation effect | |
| CN105825002A (en) | Method for modeling dynamic equivalence of wind power farm based on dynamic grey-relevancy analysis method | |
| CN115133538B (en) | Prediction control method for important load voltage model of power distribution network | |
| CN113394813B (en) | Method for calculating unit power instruction value of offshore wind farm and distributed scheduling method | |
| Hu et al. | An active primary frequency regulation strategy for grid integrated wind farms based on model predictive control | |
| CN109816261A (en) | Marine half direct drive wind power field equivalence method, system and device | |
| Hu et al. | Inertial response identification algorithm for the development of dynamic equivalent model of DFIG-based wind power plant | |
| CN107846034A (en) | A kind of energy storage for stabilizing wind power plant climbing rate goes out force control method | |
| CN117375068A (en) | Nonlinear affine-based wind farm voltage state calculation and optimization method and computer-readable medium | |
| CN117060430A (en) | Improved LSTM network-based grid-structured wind power frequency modulation prediction method | |
| CN118100156A (en) | On-line evaluation method for wind power plant frequency supporting capability based on Koopman operator | |
| CN115207941A (en) | Method for evaluating inertia level of power system | |
| CN115800260A (en) | Power system frequency response modeling method containing high-proportion new energy | |
| CN115241922A (en) | Method and system for determining unit combination of wind power plant participating in power grid frequency support | |
| CN114597908A (en) | Wind power plant reactive power control optimization method and device considering voltage distribution of current collection line | |
| CN110649604B (en) | Damping control method suitable for random fluctuation of new energy | |
| CN113991647A (en) | Frequency response capacity planning-oriented random production simulation method for power system | |
| CN113013909A (en) | Energy storage capacity improvement method based on stabilizing traction | |
| Xu et al. | MPC-BASED ACTIVE POWER COORDINATED CONTROL OF WIND POWER IN SENDING-END POWER GRID | |
| CN116388231B (en) | Wind power cluster aggregation equivalence method based on frequency and wind speed | |
| CN116599163B (en) | High-reliability wind farm power control system based on frequency modulation control | |
| CN118336762B (en) | Frequency modulation optimization method, system, terminal and storage medium for wind-storage combined system |
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 |