CN113410833A - Method for identifying active frequency response control coherent cluster - Google Patents

Method for identifying active frequency response control coherent cluster Download PDF

Info

Publication number
CN113410833A
CN113410833A CN202110570374.7A CN202110570374A CN113410833A CN 113410833 A CN113410833 A CN 113410833A CN 202110570374 A CN202110570374 A CN 202110570374A CN 113410833 A CN113410833 A CN 113410833A
Authority
CN
China
Prior art keywords
coherent
frequency response
cluster
control
response control
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
CN202110570374.7A
Other languages
Chinese (zh)
Other versions
CN113410833B (en
Inventor
晋萃萃
甘志勇
刘力卿
段明辉
姚创
赵琦
董鹏
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
State Grid Tianjin Electric Power Co Ltd
Electric Power Research Institute of State Grid Tianjin Electric Power Co Ltd
Original Assignee
State Grid Corp of China SGCC
State Grid Tianjin Electric Power Co Ltd
Electric Power Research Institute of State Grid Tianjin 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, State Grid Tianjin Electric Power Co Ltd, Electric Power Research Institute of State Grid Tianjin Electric Power Co Ltd filed Critical State Grid Corp of China SGCC
Priority to CN202110570374.7A priority Critical patent/CN113410833B/en
Publication of CN113410833A publication Critical patent/CN113410833A/en
Application granted granted Critical
Publication of CN113410833B publication Critical patent/CN113410833B/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
    • 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
    • H02J3/002—Flicker reduction, e.g. compensation of flicker introduced by non-linear load
    • 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
    • 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
    • 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
    • H02J2203/10—Power transmission or distribution systems management focussing at grid-level, e.g. load flow analysis, node profile computation, meshed network optimisation, active network management or spinning reserve management

Landscapes

  • Engineering & Computer Science (AREA)
  • Power Engineering (AREA)
  • Physics & Mathematics (AREA)
  • Nonlinear Science (AREA)
  • Supply And Distribution Of Alternating Current (AREA)

Abstract

The invention relates to an identification method of an active frequency response control coherent machine group, which is based on the angle that the machine group frequency track information provided by WAMS can reflect the machine group coherence of the active frequency response control of a power system, and utilizes a sequence minimum optimization algorithm to obtain a data sample minimum surrounding hypersphere of a Gaussian kernel function under different width coefficients so as to obtain a plurality of grouping modes, and provides a more reasonable machine group grouping mode for the active frequency response control according to the density evaluation index of the coherent machine group. The active frequency response control in the invention is suitable for clustering by taking the frequency as the clustering basis, considers the influence of fault positions, operation modes, model parameters and the like on clustering results, and has the advantages of less parameters, small calculated amount, strong operability and convenient application. Meanwhile, the support vector clustering method is feasible and effective in identification of the active frequency response control coherent cluster, can provide various coherent clustering selections for the active frequency response control, and can improve the calculation efficiency and reduce the control cost on the premise of ensuring the control precision as much as possible.

Description

Method for identifying active frequency response control coherent cluster
Technical Field
The invention belongs to the technical field of power systems, and particularly relates to an identification method for an active frequency response control coherent machine group.
Background
Along with the continuous enlargement of the scale of an extra-high voltage alternating current-direct current hybrid power grid and the increase of the grid merging quantity of a renewable energy source unit year by year, the active power balance problem of a power system is increasingly prominent, the frequency stability control situation is very severe, the system frequency safety is difficult to maintain under the condition of high-power deficiency only by using the traditional frequency response control, and the active frequency response control is required.
The active frequency response control is different from the traditional frequency response control, comprises two links of off-line analysis and on-line application, needs to carry out centralized feedforward control according to off-line predetermined parameters, and has complex calculation process and great control implementation difficulty. Because the establishment of the active frequency response control strategy is related to various factors such as control targets, local frequency differences, adjustment characteristics and the like of each unit in the system, if the control strategy is established for a single unit and a strategy mode of one unit is adopted, the problem of rapid processing of mass data can be solved, the implementation complexity is extremely high, and the online application is difficult.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provides an identification method of an active frequency response control coherent machine group, which can improve the calculation efficiency and reduce the control cost on the premise of ensuring the control precision as much as possible.
The technical problem to be solved by the invention is realized by adopting the following technical scheme:
an identification method for an active frequency response control coherent machine group comprises the following steps:
step 1, determining active frequency response control coherent identification input;
step 2, controlling coherent identification input according to the determined active frequency response, and calculating a minimum surrounding hypersphere of a high-dimensional feature space data sample;
step 3, selecting a coherent cluster distribution and grouping mode;
and 4, evaluating the active frequency response control clustering effect according to the result of the step.
Moreover, the specific implementation method of the step 1 is as follows: the frequency is used as an active frequency response control coherence identification input.
Moreover, the specific implementation method of the step 2 is as follows:
there is a non-linear mapping Φ: rd→ H will be system generator set data sample Δ fiMapping to the high-dimensional feature space H, the data sample phi (Δ f) can be obtained in the high-dimensional feature space Hi) If all data samples Φ (Δ f) are enclosed in the high-dimensional feature space Hi) Is R, the minimum enclosing hypersphere radius is:
Figure BDA0003082371880000011
wherein PgP is Euclidean norm, a is the center of the hypersphere sphere, C is a penalty factor, ξiIs a relaxation variable, and xii≥0；
The dual function of the corresponding Lagrangian function is as follows:
Figure BDA0003082371880000012
wherein N is the number of generator sets in the system, K is a Gaussian kernel function, and betaiAnd betajΔ f being Lagrangian multiplierjA data sample of the generator set j;
any data sample phi (delta f) in high-dimensional feature spacei) The distance to the center of sphere a satisfies:
Figure BDA0003082371880000021
wherein, betakIs Lagrange multiplier, Δ fkA data sample of the generator set k;
will optimize the resulting variable betaiSubstituting the values to obtain all data samples phi (delta f) surrounded in the high-dimensional characteristic space Hi) The minimum enclosed hypersphere radius of (c) is:
R＝{R(Φ(Δfi))|Φ(Δfi) Is the support vector }
The minimum bounding hypersphere boundary satisfies:
{x|R(x)＝R}。
moreover, the specific implementation method of step 3 is as follows: if the generator set i and the generator set j belong to two different coherent machine groups respectively, phi (delta f) is connected in the high-dimensional characteristic space Hi)、Φ(Δfj) Must have a point y outside the minimum bounding hypersphere, satisfy r (y)>And R, enabling an adjacency matrix A representing the adjacency relation among generator groups to satisfy:
Figure BDA0003082371880000022
wherein, phi (Δ f)i) And Φ (Δ f)j) Are all the samples of the data, and,
the density evaluation indexes of the coherent cluster are as follows:
Figure BDA0003082371880000023
wherein m is a coherent clusterNumber, dinterFor coherent cluster BiAnd BjInter-class distance between, BiFor the coherent cluster, V (m) is a density evaluation index of the coherent cluster, and the smaller the value of V (m), the better the cluster result of the coherent cluster of the unit,
in a high-dimensional characteristic space, along with the increase of the width coefficient of a Gaussian kernel function, the number of coherent clusters in the system is increased, and the evaluation index of the optimal cluster density of the power system meets the following requirements:
Figure BDA0003082371880000024
Figure BDA0003082371880000025
q＝qmin+qstep
wherein q isminIs the minimum value of the kernel function width coefficient, qstepIs the kernel width coefficient increment, q is the kernel width coefficient, VoFor the evaluation index of the optimal cluster density of the system, V (m) is the evaluation index of the cluster density of the coherent clusters, N is the number of the generator sets in the system, and delta fjIs a data sample for genset j.
And the density evaluation index of the coherent cluster is as follows: the similarity of the frequency response dynamic process between the power generating units in the same cluster is higher, and the similarity of the power generating units in different clusters is lower; the correlation between the generator groups in the same cluster is strong, and the correlation between the generator groups in different clusters is weak.
Moreover, the specific implementation method of the step 4 is as follows: the system adopts group control, each unit in the system performs feedforward control according to the prediction parameters of the cluster at the frequency reduction stage, and performs feedback control according to the local frequency difference at the frequency recovery stage:
Figure BDA0003082371880000031
wherein the content of the first and second substances,Δflclustermaximum frequency difference value c of all generator sets in the coherent machine group under the control of traditional frequency responseAParameter c for controlBParameter Δ f for controllLocal frequency difference, t, being the undisturbed pointnadirIs the time to nadir.
The invention has the advantages and positive effects that:
1. according to the invention, from the aspect that the unit frequency track information provided by the WAMS can reflect the unit coherence of the active frequency response control of the power system, the minimum surrounding hypersphere of the data sample of the Gaussian kernel function under different width coefficients is obtained by means of the sequence minimum optimization algorithm to obtain a plurality of grouping modes, and a more reasonable unit grouping mode is provided for the active frequency response control according to the coherence group density evaluation index. The active frequency response control of the invention is more suitable by taking the frequency as the clustering basis, can comprehensively consider the influence of fault position, operation mode, model parameters and the like on the clustering result, has the characteristics of less parameters, small calculated amount and the like, has strong operability and is convenient for practical application.
2. The support vector clustering method is feasible and effective in identification of the active frequency response control coherent cluster, can provide various coherent grouping selections for the active frequency response control, and can improve the calculation efficiency and reduce the control cost on the premise of ensuring the control precision as much as possible.
3. The invention uses the active frequency response grouping control, not only has little influence on the lowest point of the transient frequency of the system, but also can control the highest transient frequency of the unit far away from the disturbance point within the frequency safety constraint range of the power system, thereby preventing the system from generating new frequency safety and stability problems.
Drawings
FIG. 1 is a diagram of a system architecture for improving the system architecture of a new England 10 machine 39 node in accordance with an embodiment of the present invention;
FIG. 2 is a system frequency trace graph after a node 30 is disturbed according to an embodiment of the present invention;
FIG. 3 is a diagram illustrating the system clustering result when the cluster number is 4 according to an embodiment of the present invention;
fig. 4 is a graph comparing the active frequency response clustering control effect of the improved new england level 10 39-node system in the embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings.
Aiming at the problems in the background art, the invention is based on the angle that the unit frequency track information provided by the WAMS can reflect the unit coherence of the active frequency response control of the power system, and the minimum surrounding hypersphere of the data sample of the Gaussian kernel function under different width coefficients is obtained by means of the sequence minimum optimization algorithm to obtain a plurality of grouping modes, so that a more reasonable unit grouping mode is provided for the active frequency response control according to the coherence group density evaluation index.
An identification method for an active frequency response control coherent machine group comprises the following steps:
step 1, determining active frequency response control coherent identification input.
The trace information such as the frequency and the power angle of the generator set collected by the PMU can reflect main factors influencing the unit coherence, such as system model parameters, an operation mode, disturbance positions and the like. During active frequency response control, when a unit far away from a disturbance point supports a unit near the disturbance point in power, the highest frequency of the unit far away from the disturbance point needs to be focused on so as to prevent the operating frequency of the unit far away from the disturbance point from exceeding the safety upper limit of the power system frequency, thereby causing a new frequency safety and stability problem.
Considering that the generator sets with approximately consistent frequency dynamic processes can carry out unified management control on the aspects of frequency safety constraint setting, control strategy switching termination and the like, the invention takes the frequency as the active frequency response control coherence identification input so as to fully utilize the frequency response capability of each set in the system on the premise of ensuring the operation safety of the system, and reduce or even avoid the phenomenon that part of the sets in a coherence group are close to the upper frequency safety limit and can not continuously carry out the active frequency response control, and the other sets still do not fully exert the frequency response capability.
And 2, controlling coherent identification input according to the determined active frequency response, and calculating the minimum surrounding hypersphere of the high-dimensional feature space data sample.
There is a non-linear mapping Φ: rd→ H will be system generator set data sample Δ fiMapping to the high-dimensional feature space H, the data sample phi (Δ f) can be obtained in the high-dimensional feature space Hi) If all data samples Φ (Δ f) are enclosed in the high-dimensional feature space Hi) Is R, the minimum bounding hypersphere radius solution problem can be transformed into an optimization problem:
Figure BDA0003082371880000041
wherein PgP is Euclidean norm, a is the center of the hypersphere sphere, C is a penalty factor, ξiIs a relaxation variable, and xii≥0；
The dual function of the corresponding Lagrangian function is as follows:
Figure BDA0003082371880000042
wherein N is the number of generator sets in the system, K is a Gaussian kernel function, and betaiAnd betajΔ f being Lagrangian multiplierjIs a data sample for genset j.
Any data sample phi (delta f) in high-dimensional feature spacei) The distance to the center of sphere a satisfies:
Figure BDA0003082371880000043
wherein, betakIs Lagrange multiplier, Δ fkIs a data sample of genset k.
Will optimize the resulting variable betaiSubstituting the value of (d) into the data sample phi (Δ f)i) Obtaining a distance formula from the sphere center a to surround all data samples phi (delta f) in the high-dimensional feature space Hi) The minimum enclosed hypersphere radius of (c) is:
R＝{R(Φ(Δfi))|Φ(Δfi) Is the support vector }
The minimum bounding hypersphere boundary satisfies:
{x|R(x)＝R}。
and 3, selecting a coherent cluster distribution and grouping mode.
If the generator set i and the generator set j belong to two different coherent machine groups respectively, phi (delta f) is connected in the high-dimensional characteristic space Hi)、Φ(Δfj) Must have a point y outside the minimum bounding hypersphere, satisfy r (y)>And R, enabling an adjacency matrix A representing the adjacency relation among generator groups to satisfy:
Figure BDA0003082371880000051
wherein, phi (Δ f)i) And Φ (Δ f)j) Are all data samples.
The density evaluation index of the coherent cluster facing the active frequency response control of the power system mainly comprises two contents: firstly, the similarity of the frequency response dynamic process between the power generation units in the same cluster is higher, and the similarity between the power generation units in different clusters is lower; secondly, the correlation between the generator groups in the same cluster is strong, and the correlation between the generator groups in different clusters is weak. According to the density evaluation index of the coherent cluster, the density evaluation index of the coherent cluster is defined as:
Figure BDA0003082371880000052
wherein m is the number of coherent clusters, BiFor coherent clusters, dinter(Bi) Representing coherent fleet BiAnd BjThe inter-class distance V (m) is an evaluation index of the density of coherent clusters, and the smaller the value of V (m), the better the coherent clustering result of the clusters,
in a high-dimensional characteristic space, along with the increase of the width coefficient of a Gaussian kernel function, the number of coherent clusters in the system is increased, and the evaluation index of the optimal cluster density of the power system meets the following requirements:
Figure BDA0003082371880000053
Figure BDA0003082371880000054
q＝qmin+qstep
wherein q isminIs the minimum value of the kernel function width coefficient, qstepIs the kernel width coefficient increment, q is the kernel width coefficient, VoFor the evaluation index of the optimal cluster density of the system, V (m) is the evaluation index of the cluster density of the coherent clusters, N is the number of the generator sets in the system, and delta fjIs a data sample for genset j.
And 4, evaluating the active frequency response control clustering effect according to the result of the step.
Suppose that the local frequency difference of disturbance point in the system is delta frTime to lowest point is tnadirThe control criterion is cAThe local frequency difference at the non-disturbance point is DeltaflThe control criterion is cBIf the system does not adopt the group control, each unit in the system performs feedforward control according to the system prediction parameters at the frequency reduction stage and performs feedback control according to the local frequency difference at the frequency recovery stage:
Figure BDA0003082371880000061
in this control mode, although all units in the system can fully exert their frequency response capability, and the lowest point of the transient frequency of the system is maximally improved, the frequency stability of the units far from the disturbance point is affected. On one hand, the frequency of the unit far away from the disturbance point can be reversely adjusted, so that the highest frequency of the unit is close to or even exceeds the safety upper limit of the system operation frequency, and a new frequency stability problem is caused; on the other hand, the frequency oscillation of the unit far away from the disturbance point in the frequency recovery stage is serious.
In order to solve the problems, when the active frequency response control is carried out, the system adopts the grouping control to reduce the calculation complexity and improve the control efficiency. If the system adopts the grouping control, each unit in the system carries out feedforward control according to the prediction parameters of the cluster at the frequency reduction stage, and carries out feedback control according to the local frequency difference at the frequency recovery stage:
Figure BDA0003082371880000062
wherein, Δ flclusterMaximum frequency difference value c of all generator sets in the coherent machine group under the control of traditional frequency responseAParameter c for controlBParameter Δ f for controllLocal frequency difference, t, being the undisturbed pointnadirIs the time to nadir.
According to the identification method of the active frequency response control coherent machine group, the improved 39-node system of the new England 10 machine is analyzed to verify the functional effect of the invention.
Assuming stable operation of the system before disturbance, the rated operating frequency and the low frequency load shedding threshold are 50Hz and 49.5Hz, respectively. The clustering parameters of the method provided by the invention meet the following requirements: penalty factor C is 1 and slack variable ξ is 0.2.
The improved new england 10-machine 39-node system comprises 10 generator sets, 39 nodes and 34 transmission lines, and a system topological diagram is shown in figure 1. When t is 2s, the node 30 is disturbed, the power shortage is 0.135p.u., and the frequency dynamic locus of each generator node in the system is shown in fig. 2.
As shown in FIG. 2, after the disturbance occurs, the lowest point of the transient frequency of the system is-0.5175 Hz, low-frequency load shedding has occurred, and active frequency response control needs to be adopted. Because the difference between the frequency dynamic tracks of each generator node is mainly reflected in 2-15 s, the frequency sampling value in the time period can be used for constructing a data sample for each generator node. Meanwhile, the system obtained by using the present invention has 7 coherent grouping options, as shown in Table 1.
Watch 1
Figure BDA0003082371880000063
When Vo is 0.1501, the coherent clustering result of the unit facing the active frequency response control is most reasonable, and the number of clusters is 3, as shown in fig. 3.
When the number of clusters is 3, the active frequency response control effect before and after the system clustering is shown in fig. 4. As can be seen from the simulation results, before clustering, the transient lowest frequency and the transient highest frequency of the active frequency response control of the system are-0.4954 Hz and 0.2249Hz, respectively; after clustering, the transient lowest frequency and the transient highest frequency of the active frequency response control of the system are-0.4952 Hz and 0.1866Hz, respectively. Therefore, the system not only has little influence on the lowest frequency of the transient state of the system by implementing active frequency response clustering control, but also can control the highest frequency of the transient state of the system within a safety constraint range, thereby avoiding the problem of new frequency safety and stability of the system.
It should be emphasized that the embodiments described herein are illustrative rather than restrictive, and thus the present invention is not limited to the embodiments described in the detailed description, but also includes other embodiments that can be derived from the technical solutions of the present invention by those skilled in the art.

Claims (6)

1. A method for identifying an active frequency response control coherent cluster is characterized in that: the method comprises the following steps:
step 1, determining active frequency response control coherent identification input;
step 2, controlling coherent identification input according to the determined active frequency response, and calculating a minimum surrounding hypersphere of a high-dimensional feature space data sample;
step 3, selecting a coherent cluster distribution and grouping mode;
and 4, evaluating the active frequency response control clustering effect according to the result of the step.
2. The method of claim 1, wherein the method comprises: the specific implementation method of the step 1 comprises the following steps: the frequency is used as an active frequency response control coherence identification input.
3. The method of claim 1, wherein the method comprises: the specific implementation method of the step 2 comprises the following steps:
there is a non-linear mapping Φ: rd→ H will be system generator set data sample Δ fiMapping to the high-dimensional feature space H, the data sample phi (Δ f) can be obtained in the high-dimensional feature space Hi) If all data samples Φ (Δ f) are enclosed in the high-dimensional feature space Hi) Is R, the minimum enclosing hypersphere radius is:
Figure FDA0003082371870000011
wherein PgP is Euclidean norm, a is the center of the hypersphere sphere, C is a penalty factor, ξiIs a relaxation variable, and xii≥0；
The dual function of the corresponding Lagrangian function is as follows:
Figure FDA0003082371870000012
wherein N is the number of generator sets in the system, K is a Gaussian kernel function, and betaiAnd betajΔ f being Lagrangian multiplierjA data sample of the generator set j;
any data sample phi (delta f) in high-dimensional feature spacei) The distance to the center of sphere a satisfies:
Figure FDA0003082371870000013
wherein, betakIs Lagrange multiplier, Δ fkA data sample of the generator set k;
will optimize the resulting variable betaiSubstituting the values to obtain all data samples phi (delta f) surrounded in the high-dimensional characteristic space Hi) The minimum enclosed hypersphere radius of (c) is:
R＝{R(Φ(Δfi))|Φ(Δfi) Is the support vector }
The minimum bounding hypersphere boundary satisfies:
{x|R(x)＝R}。
4. the method of claim 1, wherein the method comprises: the specific implementation method of the step 3 is as follows: if the generator set i and the generator set j belong to two different coherent machine groups respectively, phi (delta f) is connected in the high-dimensional characteristic space Hi)、Φ(Δfj) Must have a point y outside the minimum bounding hypersphere, satisfy r (y)>And R, enabling an adjacency matrix A representing the adjacency relation among generator groups to satisfy:
Figure FDA0003082371870000021
wherein, phi (Δ f)i) And Φ (Δ f)j) Are all the samples of the data, and,
the density evaluation indexes of the coherent cluster are as follows:
Figure FDA0003082371870000022
wherein m is the number of coherent clusters, dinterFor coherent cluster BiAnd BjInter-class distance between, BiFor the coherent cluster, V (m) is a density evaluation index of the coherent cluster, and the smaller the value of V (m), the better the cluster result of the coherent cluster of the unit,
in a high-dimensional characteristic space, along with the increase of the width coefficient of a Gaussian kernel function, the number of coherent clusters in the system is increased, and the evaluation index of the optimal cluster density of the power system meets the following requirements:
Figure FDA0003082371870000023
Figure FDA0003082371870000024
q＝qmin+qstep
wherein q isminIs the minimum value of the kernel function width coefficient, qstepIs the kernel width coefficient increment, q is the kernel width coefficient, VoFor the evaluation index of the optimal cluster density of the system, V (m) is the evaluation index of the cluster density of the coherent clusters, N is the number of the generator sets in the system, and delta fjIs a data sample for genset j.
5. The method of claim 4, wherein the method comprises: the density evaluation indexes of the coherent cluster are as follows: the similarity of the frequency response dynamic process between the power generating units in the same cluster is higher, and the similarity of the power generating units in different clusters is lower; the correlation between the generator groups in the same cluster is strong, and the correlation between the generator groups in different clusters is weak.
6. The method of claim 1, wherein the method comprises: the specific implementation method of the step 4 comprises the following steps: the system adopts group control, each unit in the system performs feedforward control according to the prediction parameters of the cluster at the frequency reduction stage, and performs feedback control according to the local frequency difference at the frequency recovery stage:
Figure FDA0003082371870000025
wherein, Δ flclusterMaximum frequency difference value c of all generator sets in the coherent machine group under the control of traditional frequency responseAParameter c for controlBParameter Δ f for controllLocal frequency difference, t, being the undisturbed pointnadirIs the time to nadir.
CN202110570374.7A 2021-05-25 2021-05-25 Identification method for active frequency response control coherent group Active CN113410833B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110570374.7A CN113410833B (en) 2021-05-25 2021-05-25 Identification method for active frequency response control coherent group

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110570374.7A CN113410833B (en) 2021-05-25 2021-05-25 Identification method for active frequency response control coherent group

Publications (2)

Publication Number Publication Date
CN113410833A true CN113410833A (en) 2021-09-17
CN113410833B CN113410833B (en) 2024-04-19

Family

ID=77674747

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110570374.7A Active CN113410833B (en) 2021-05-25 2021-05-25 Identification method for active frequency response control coherent group

Country Status (1)

Country Link
CN (1) CN113410833B (en)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2004016218A2 (en) * 2002-08-15 2004-02-26 Pacific Edge Biotechnology, Ltd. Medical decision support systems utilizing gene expression and clinical information and method for use
CN108288858A (en) * 2017-12-18 2018-07-17 国网辽宁省电力有限公司经济技术研究院 Main passive frequency response method for handover control
WO2019161589A1 (en) * 2018-02-24 2019-08-29 大连理工大学 Real-time tracking method for structural modal parameter
CN110277783A (en) * 2019-05-30 2019-09-24 国网河南省电力公司电力科学研究院 A kind of online Active Splitting Control and system of electric system

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2004016218A2 (en) * 2002-08-15 2004-02-26 Pacific Edge Biotechnology, Ltd. Medical decision support systems utilizing gene expression and clinical information and method for use
CN108288858A (en) * 2017-12-18 2018-07-17 国网辽宁省电力有限公司经济技术研究院 Main passive frequency response method for handover control
WO2019161589A1 (en) * 2018-02-24 2019-08-29 大连理工大学 Real-time tracking method for structural modal parameter
CN110277783A (en) * 2019-05-30 2019-09-24 国网河南省电力公司电力科学研究院 A kind of online Active Splitting Control and system of electric system

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
宋洪磊;吴俊勇;郝亮亮;冀鲁豫;: "基于WAMS和改进拉普拉斯特征映射的同调机群在线识别", 电网技术, no. 08 *
李卫东;晋萃萃;温可瑞;申家锴;刘柳;: "大功率缺失下主动频率响应控制初探", 电力系统自动化, no. 08 *

Also Published As

Publication number Publication date
CN113410833B (en) 2024-04-19

Similar Documents

Publication Publication Date Title
Sun et al. Model predictive control and improved low-pass filtering strategies based on wind power fluctuation mitigation
Wang et al. Underfrequency load shedding scheme for islanded microgrids considering objective and subjective weight of loads
CN114726009B (en) Wind power plant group reactive power hierarchical optimization control method and system considering power prediction
Guo et al. Two‐stage optimal MPC for hybrid energy storage operation to enable smooth wind power integration
CN112380694A (en) Power distribution network optimization planning method based on differential reliability requirements
CN113516278A (en) Active power distribution network multi-time scale active and reactive power coordinated optimization scheduling method and system
Syed et al. An Intelligent Model Predictive Control Strategy for Stable Solar‐Wind Renewable Power Dispatch Coupled with Hydrogen Electrolyzer and Battery Energy Storage
Beus et al. A model predictive control approach to operation optimization of an ultracapacitor bank for frequency control
Abdulwahid Artificial intelligence-based control techniques for hvdc systems
CN109888824B (en) Photovoltaic grid-connected inversion control method based on predictive control
CN113410833A (en) Method for identifying active frequency response control coherent cluster
Cui et al. Building a prediction model of solar power generation based on improved Grey Markov Chain
CN115051360A (en) Online computing method and device for operation risk of electric power system based on integrated knowledge migration
Liu et al. Equivalent modeling of distributed photovoltaic power station clusters based on deep belief network
Qu et al. Active and Reactive Power Coordinated Optimal Dispatch in Active Distribution Network Considering Spatial-temporal Correlation of Wind Power
Tang et al. Static voltage stability margin prediction of island microgrid based on Tri-Training-Lasso-BP network
Zhu et al. Frequency Stability Control Method of AC/DC Power System Based on Convolutional Neural Network
Sahoo et al. Forecasting Tariff Rates and Enhancing Power Quality in Microgrids: The Synergistic Role of LSTM and UPQC
Guo et al. Voltage stability control method based on comprehensive principal component analysis and improved neural network
Ma et al. Impact and solution of pv power fluctuations on operation of hybrid ac/dc microgrid
Hong et al. Harmonic Prediction Algorithm Based on Optimized Neural Network
CN109921514A (en) Method for power quality controlling decision support
CN113852123B (en) Automatic voltage control method and device for power system, electronic equipment and storage medium
Wang et al. Equivalent Modeling of PV Cluster Dynamic Process Based on K-medoids Clustering
Lin et al. Emergency Control of Power Grid under Topology Changes Based on Graph Reinforcement Learning

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