CN109829599B - Cluster division method and device for power distribution network based on high-proportion renewable energy - Google Patents
Cluster division method and device for power distribution network based on high-proportion renewable energy Download PDFInfo
- Publication number
- CN109829599B CN109829599B CN201811397608.7A CN201811397608A CN109829599B CN 109829599 B CN109829599 B CN 109829599B CN 201811397608 A CN201811397608 A CN 201811397608A CN 109829599 B CN109829599 B CN 109829599B
- Authority
- CN
- China
- Prior art keywords
- node
- current individual
- cluster
- power
- serial number
- 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.)
- Active
Links
Images
Abstract
The invention discloses a cluster division method of a power distribution network based on high-proportion renewable energy, which comprises the following steps: 1) abstracting the power distribution network into a network formed by connecting nodes and edges; 2) randomly changing a plurality of first preset values in the network into second preset values, and acquiring a plurality of individuals corresponding to the changed network; 3) acquiring the cluster performance index of the current individual according to the modularity index and the surplus power index; 4) judging whether the iteration is converged; 5) if so, taking the changed network connection architecture as a target connection architecture; 6) and if not, selecting individuals, crossing the individuals and performing variation processing on the individuals, taking the individuals corresponding to the processed network as the current individuals, and returning to execute the step 4) until the target connection architecture is obtained. The invention discloses a cluster dividing device of a power distribution network based on high-proportion renewable energy. By applying the embodiment of the invention, the renewable energy consumption capability of the system can be improved through the cluster.
Description
Technical Field
The invention relates to a cluster division method and a cluster division device, in particular to a cluster division method and a cluster division device of a power distribution network based on high-proportion renewable energy.
Background
Human concerns about fossil energy depletion, energy safety and environmental deterioration have led to the continuous and rapid development of renewable energy, and a power system scenario with a high proportion of renewable energy for power generation is attracting much attention. Particularly, in a power distribution network in remote areas of China, along with further strengthening of the national poverty-relief policy of new energy, a large amount of distributed renewable energy is connected into the power distribution network, and even in the case that the permeability of part of areas is more than 100%, the situation can give a great influence to the safety and stability of a local power distribution network, and the situation is mainly reflected in the aspects of voltage overlimit, power over-voltage grade reverse transmission and the like. Therefore, how to manage new energy access is an urgent technical problem to be solved.
At present, when a cluster control mode is generally used for solving the problem of management difficulty caused by small unit capacity, large quantity, dispersed geographic positions and the like of a large-scale renewable energy source access power distribution network, the cluster control mode is more and more widely applied due to the characteristics of cooperation of unique weak coupling among clusters, division of labor and strong connection in the clusters, and the advantages that compared with an on-site control cluster, information interaction with other regions of a system can be realized to achieve optimal control, and compared with a global control cluster, the response speed is higher. The principle of cluster control is as follows: the target power grid is divided into a plurality of sub-areas, each area is a cluster and is externally represented as a whole, and the target power grid is controlled by a single instruction and is convenient to dispatch and manage; all nodes in the cluster cooperate with each other to complete a common target, and the cooperation capability among the nodes is effectively exerted. As can be seen from the existing literature, in a power system containing renewable energy, the application scenario of a cluster already covers the operation control field of the system, and the cluster is divided based on the coupling relationship between voltage change between nodes and active and reactive change, and the common dividing method includes: based on the concept of voltage sensitivity coefficient, the electrical coupling strength between nodes is expressed, and the method is used for cluster division of voltage control of the power distribution network; according to the relation that the node voltage changes along with the active and reactive changes, a sensitivity matrix is constructed to carry out cluster division on a control layer of a power distribution system; in order to simplify the operation and control of a power system, an electrical distance is constructed according to the relation between active power and voltage phase angle sensitivity among nodes, and a network is clustered and partitioned according to intra-cluster distance and inter-cluster distance indexes.
From the above, the cluster has obvious advantages in solving the problem of difficult management caused by the access of large-scale renewable energy sources to the power distribution network. However, the power characteristics of each node cannot be reflected by the division of the power distribution network by the electrical coupling among the nodes in the prior art, so that the cluster division of the power distribution network in the prior art is not accurate enough.
Disclosure of Invention
The invention aims to provide a method and a device for cluster division of a power distribution network based on high-proportion renewable energy sources, so as to solve the technical problem that the cluster division of the power distribution network in the prior art is not accurate enough.
The invention solves the technical problems through the following technical scheme:
the embodiment of the invention provides a cluster division method of a power distribution network based on high-proportion renewable energy, which comprises the following steps:
1) abstracting a power distribution network to be subjected to cluster division into a network formed by connecting nodes and edges, using a first preset value to represent a connection relation between two nodes which are connected with each other in the network, and using a second preset value to represent a connection relation between two nodes which are not connected with each other in the network, wherein each power station in the power distribution network is abstracted into the nodes, and connecting lines between the power stations are abstracted into the edges;
2) Randomly and respectively changing a plurality of first preset values in the network into second preset values, acquiring a plurality of individuals corresponding to the changed network, and taking the individuals as current individuals, wherein each individual comprises at least one cluster, and the cluster is a sub-network formed by connecting a plurality of nodes in the network;
3) aiming at each current individual, acquiring a modularity index of the current individual according to the electrical connection closeness degree between nodes in the current individual, acquiring a surplus electric quantity index of the current individual according to a power value of each node in the current individual and the power regulation capacity of the current individual for energy storage, and acquiring a cluster performance index of the current individual according to the modularity index and the surplus electric quantity index;
4) judging whether the minimum value in the cluster performance indexes respectively corresponding to the current individuals is smaller than a third preset threshold value or whether the iteration times reach a fourth preset threshold value;
5) if so, taking the network connection architecture represented by the current individual corresponding to the minimum value of the cluster performance index value as a target connection architecture;
6) and if not, utilizing a genetic algorithm to perform individual selection, individual crossing and individual variation processing on the population, taking the individual corresponding to the processed network as the current individual, and returning to execute the step 4) until a target connection architecture is obtained.
Optionally, the obtaining the modularity index of the current individual according to the closeness degree of the electrical connection between the nodes in the current individual includes:
by means of the formula (I) and (II),
calculating an electrical distance of the current individual, wherein,
by means of the formula (I) and (II),
calculating the electrical distance between node i and node j, wherein,
LijIs the electrical distance between node i and node j; x is the number of the time sections, and t belongs to x; sigma is a summation function;
using the formula, eij＝1-Lij(l), calculating weights of edges between nodes, wherein,
eijthe weight of the edge between the ith node and the jth node; l isijIs the electrical distance between node i and node j; max (L) is the maximum value of the weights in all edges of the connection;
by means of the formula (I) and (II),
obtaining a modularity index for the current individual, wherein,
rho is a modularity index in the current individual; m is the sum of the weights of all edges in the current individual; e.g. of the typeijThe weight of the edge between the ith node and the jth node; k is a radical ofiIs the sum of the weights of all edges connected to node i; k is a radical ofjIs the sum of the weights of all edges connected to node j; sigma (i, j) is a characteristic parameter of whether the node i and the node j are in the same cluster, and the value of the same cluster is 1Different from 0; i is a node serial number; j is the node sequence number.
Optionally, the obtaining, by the power adjustment capability of the current individual for energy storage, the surplus power indicator of the current individual includes:
by means of the formula (I) and (II),
calculating the net power of the cluster with the serial number c in the current individual at the moment section t, wherein,
by means of the formula (I) and (II),
calculating the power regulation capacity of the controllable load of the cluster with the serial number c in the current individual at the moment section t, wherein,
by means of the formula (I) and (II),
calculating the current individual isThe power regulation capability of the stored energy of the ith energy storage device at the moment of the section t, wherein,
by means of the formula (I) and (II),
calculating the power regulation capacity of the energy storage device of the cluster with the serial number c in the current individual at the moment section t, wherein,
by means of the formula (I) and (II),
calculating the surplus electric quantity index of the cluster with the serial number c in the current individual after power adjustment at the moment section t, wherein,
by means of the formula (I) and (II),
and calculating the surplus power index of the current individual after the consumption capacity is considered, wherein,
Optionally, the obtaining the cluster performance index of the current individual according to the modularity index and the surplus power index includes:
by means of the formula (I) and (II),
a cluster performance indicator for the current individual is calculated, wherein,
gamma is the cluster performance index of the current individual;
surplus electric quantity indexes after the consumption capacity is considered for the current individuals; omega1The weight of the surplus power index after the consumption capacity is considered for the current individual; rho is a modularity index in the current individual; omega2Is the weight of the modularity index in the current individual.
The embodiment of the invention provides a cluster dividing device of a power distribution network based on high-proportion renewable energy, which comprises:
the system comprises an abstraction module, a first control module and a second control module, wherein the abstraction module is used for abstracting a power distribution network to be subjected to cluster division into a network formed by connecting nodes and edges, expressing the connection relation between two mutually connected nodes in the network by using a first preset value, and expressing the connection relation between two mutually unconnected nodes in the network by using a second preset value, wherein each power station in the power distribution network is abstracted into the nodes, and connecting lines between the power stations are abstracted into the edges;
the first acquisition module is used for randomly and respectively changing a plurality of first preset values in the network into second preset values, acquiring a plurality of individuals corresponding to the changed network, and taking the individuals as current individuals, wherein each individual comprises at least one cluster, and the cluster is a sub-network formed by connecting a plurality of nodes in the network;
A second obtaining module, configured to, for each current individual, obtain a modularity index of the current individual according to a degree of closeness of electrical connection between nodes in the current individual, obtain an excess power index of the current individual according to a power value of each node in the current individual and a power adjustment capability of energy storage of the current individual, and obtain a clustering performance index of the current individual according to the modularity index and the excess power index;
the judging module is used for judging whether the minimum value in the cluster performance indexes respectively corresponding to the current individuals is smaller than a third preset threshold or whether the iteration times reach a fourth preset threshold;
the first setting module is used for taking the network connection architecture represented by the current individual corresponding to the minimum value of the cluster performance index value as a target connection architecture under the condition that the judgment result of the judgment module is yes;
and the second setting module is used for selecting individuals, crossing among the individuals and performing variation processing on the individuals on the population by using a genetic algorithm under the condition that the judgment result of the judgment module is negative, and returning the processed individuals corresponding to the network as the current individuals to the triggering judgment module until the target connection architecture is obtained.
Optionally, the second obtaining module is configured to:
by means of the formula (I) and (II),
calculating an electrical distance of the current individual, wherein,
by means of the formula (I) and (II),
calculating the electrical distance between node i and node j, wherein,
Lijis the electrical distance between node i and node j; x is the number of the time sections, and t belongs to x; sigma is a summation function;
Using the formula, eij＝1-Lij(l), calculating weights of edges between nodes, wherein,
eijthe weight of the edge between the ith node and the jth node; l isijIs the electrical distance between node i and node j; max (L) is the maximum value of the weights in all edges of the connection;
by means of the formula (I) and (II),
obtaining a modularity index for the current individual, wherein,
rho is a modularity index in the current individual; m is the sum of the weights of all edges in the current individual; e.g. of the typeijThe weight of the edge between the ith node and the jth node; k is a radical ofiIs the sum of the weights of all edges connected to node i; k is a radical ofjIs prepared by reacting withThe sum of the weights of all edges connected by node j; sigma (i, j) is a characteristic parameter of whether the node i and the node j are positioned in the same cluster, and the value of the same cluster is 1 and is different from 0; i is a node serial number; j is the node sequence number.
Optionally, the second obtaining module is configured to:
by means of the formula (I) and (II),
calculating the net power of the cluster with the sequence number c in the current individual at the moment of section t, wherein,
By means of the formula (I) and (II),
calculating the power regulation capacity of the controllable load of the cluster with the serial number c in the current individual at the moment section t, wherein,
by means of the formula (I) and (II),
calculating the power regulation capacity of the stored energy of the ith energy storage device of the current individual at the moment section t, wherein,
by means of the formula (I) and (II),
calculating the power regulation capacity of the energy storage device of the cluster with the serial number c in the current individual at the moment section t, wherein,
by means of the formula (I) and (II),
calculating the surplus electric quantity index of the cluster with the serial number c in the current individual after power adjustment at the moment section t, wherein,
by means of the formula (I) and (II),
and calculating the surplus power index of the current individual after the consumption capacity is considered, wherein,
Optionally, the second obtaining module is configured to:
by means of the formula (I) and (II),
calculating the power regulation capacity of the current individual energy storage to obtain the cluster performance index of the current individual, wherein,
gamma is the cluster performance index of the current individual;
surplus electric quantity indexes after the consumption capacity is considered for the current individuals; omega1The weight of the surplus power index after the consumption capacity is considered for the current individual; rho is a modularity index in the current individual; omega2Is the weight of the modularity index in the current individual.
Compared with the prior art, the invention has the following advantages:
by applying the embodiment of the invention, the modularity index for community detection based on the electrical distance and the cluster surplus electric quantity index for measuring the power consumption capability of the renewable energy source are used, and the genetic algorithm is utilized to divide the distributed power supply clusters, so that the method can not ensure that the consumption capability of the system can be fully utilized in planning and later-stage operation of the divided clusters due to the lack of consideration on the power regulation capability of the system compared with the cluster division method which simply takes source-load balance as a condition in the prior art, and the renewable energy consumption capability in the system can be fully exerted.
Drawings
Fig. 1 is a schematic flow chart of a cluster division method for a power distribution network based on high-proportion renewable energy according to an embodiment of the present invention;
fig. 2 is a schematic diagram illustrating a principle of a cluster division method for a power distribution network based on high-proportion renewable energy according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram of a cluster partitioning device of a power distribution network based on high-proportion renewable energy according to an embodiment of the present invention.
Detailed Description
The following examples are given for the detailed implementation and specific operation of the present invention, but the scope of the present invention is not limited to the following examples.
The embodiment of the invention provides a cluster division method and a cluster division device for a power distribution network based on high-proportion renewable energy, and firstly introduces the cluster division method for the power distribution network based on high-proportion renewable energy provided by the embodiment of the invention.
Fig. 1 is a schematic flow chart of a cluster division method for a power distribution network based on high-proportion renewable energy according to an embodiment of the present invention; fig. 2 is a schematic diagram of a principle of a cluster division method for a power distribution network based on high-proportion renewable energy according to an embodiment of the present invention, as shown in fig. 1 and fig. 2, the method includes:
S101: abstracting a power distribution network to be subjected to cluster division into a network formed by connecting nodes and edges, using a first preset value to represent a connection relation between two nodes which are connected with each other in the network, and using a second preset value to represent a connection relation between two nodes which are not connected with each other in the network, wherein each power station in the power distribution network is abstracted into the nodes, and connecting lines between the power stations are abstracted into the edges;
specifically, the power distribution network may include a plurality of power stations, each power station may have a connection relationship, the power stations are regarded as nodes, and the connection lines between the power stations are regarded as edges between the nodes, so that a complex network formed by connecting a plurality of nodes is obtained.
S102: randomly and respectively changing a plurality of first preset values in the network into second preset values, acquiring a plurality of individuals corresponding to the changed network, and taking the individuals as current individuals, wherein each individual comprises at least one cluster, and the cluster is a sub-network formed by connecting a plurality of nodes in the network;
in practical application, for a network without cluster division, when a connection relationship exists between a node i and a node j, an original value of an edge connecting the node i and the node j is set to 1, and if the connection relationship does not exist between the node i and the node j, the original value of the edge connecting the node i and the node j is set to 0;
Then, randomly changing the original value 1 of the edges among a plurality of nodes into 0 for the first time; at this time, individual 1 was obtained; randomly changing the original value 1 of the edges among the plurality of nodes into 0 for the second time; at this time, individual 2 was obtained; and in the same way, obtaining a plurality of individuals, wherein each individual corresponds to the network structure state after the network which needs to be subjected to cluster division is divided. Each individual comprises a plurality of independent sub-networks, and each sub-network is a cluster.
S103: aiming at each current individual, acquiring a modularity index of the current individual according to the close degree of electrical connection among nodes in the current individual, acquiring a surplus electric quantity index of the current individual according to the power value of each node in the current individual and the power regulation capacity of the current individual, and acquiring a cluster performance index of the current individual according to the modularity index and the surplus electric quantity index;
(1) it is possible to use the formula,
calculating an electrical distance of the current individual, wherein,
In practical application, when the section at the moment is t, the mathematical relationship between the reactive power change unit value of any node in the current individual and the voltage change value of any corresponding node meets the following power flow equation,
Δδtthe matrix is the matrix of the power angle change increment of each node in the current individual when the time section is t; Δ VtThe matrix is a matrix of voltage change increment of each node in the current individual when the time section is t; delta PtThe matrix is the matrix of the active power change increment of each node in the current individual when the time section is t; delta Q tAnd the matrix is the reactive power change increment of each node in the current individual when the time section is t. The matrixes are n-dimensional matrixes, and n is the total number of nodes contained in all the individuals;
the power angle of each node is active and sensitive under the section of the moment tA degree coefficient matrix;
the voltage active sensitivity coefficient matrix of each node under the section of the moment t;
the voltage reactive sensitivity coefficient matrix of each node under the section of the moment t;
under the section of the moment t, the power angle reactive sensitivity coefficient matrix of each node; for example in a matrix
Row i and column j of
The unit value of the reactive power change of the node j is represented to correspond to the change value of the voltage of the node i.
(2) And the use of a formula,
calculating the electrical distance between node i and node j, wherein,
Lijis the electrical distance between node i and node j; x is the number of the time sections, and t belongs to x; and Σ is a summation function.
In practical applications, when clustering is performed, the time scale of the application may be different time lengths such as a typical day, a typical month, a typical year, and the like, for example, the typical day is a time of day as a time scale for clustering. In the embodiment of the present invention, a typical day may be taken as a time scale, for example, the step size is 1 hour, and if the typical day includes 24 hours, x is 24.
(3) Using the formula eij＝1-Lij(l), calculating weights of edges between nodes, wherein,
eijis as followsThe weight of the edge between the i node and the j node; l isijIs the electrical distance between node i and node j; max (l) is the maximum value of the weights in all edges connected to node i;
(4) and the use of a formula,
obtaining a modularity index for the current individual, wherein,
rho is the modularity index of the current individual, and the value of rho is between 0 and 1; m is the sum of the weights of all edges in the current individual; e.g. of the typeijThe weight of the edge between the ith node and the jth node; k is a radical ofiIs the sum of the weights of all edges connected to node i; k is a radical ofjIs the sum of the weights of all edges connected to node j; sigma (i, j) is a characteristic parameter of whether the node i and the node j are positioned in the same cluster; i is a node serial number; j is the node sequence number.
For example, when σ (i, j) is 1, it means that the node i and the node j are located in the same cluster; when σ (i, j) is 0, it means that the node i and the node j are not located in the same cluster.
In addition, it is possible to use a formula,
the sum of the weights of all edges in the current individual is calculated.
(5) It is possible to use the formula,
calculating the net power of the current individual at the time section t with the sequence number c, wherein,
(6) and the use of a formula,
calculating the power regulation capacity of the controllable load of the cluster with the serial number c in the current individual at the moment section t, wherein,
(7) and the use of a formula,
calculating the power regulation capacity of the stored energy of the ith energy storage device of the current individual at the moment section t, wherein,
Because the power regulation capability of the energy storage device is related to the state of charge at the current moment, the electric quantity constraint needs to be satisfied when the power regulation capability is measured:
Et≤EMaxWherein, in the step (A),
Etthe electric quantity value of the energy storage device at the time t and the initial electric quantity E of the energy storage deviceinAnd charging efficiency ηcDischarge efficiency etadIn connection with this, the present invention is,and is
The real-time discharge power of the energy storage device is obtained; eMaxThe maximum amount of charge allowed by the energy storage device during charging.
(8) And the use of a formula,
calculating the power regulation capacity of the energy storage device when the cluster with the serial number c in the current individual is in the time section t, wherein,
(9) and the use of a formula,
calculating the surplus electric quantity index of the cluster with the serial number c in the current individual after power adjustment at the moment section t, wherein,
More specifically, the surplus power of the cluster c at the time t after power adjustment
In the calculation, only when the power output is greater than the load demand in the time domain, that is to say
The power regulation is performed under the condition(s). Meanwhile, in consideration of the limitation of scheduling of load side demand response, the power regulation capability of the controllable load is firstly utilized when power regulation is carried out
Adjusting if the surplus electric quantity index is not met
Power regulation capacity by stored energy
Is adjusted when
Sometimes still not satisfied
Energy-saving power regulation by stored energy
At its maximum value PcPower regulation is performed.
(10) And the use of a formula,
and calculating the surplus electric quantity index of the current individual after the consumption capacity is considered, wherein,
(11) And the use of a formula,
calculating the power regulation capacity of the current individual energy storage to obtain the cluster performance index of the current individual, wherein,
gamma is the cluster performance index of the current individual;
surplus electric quantity indexes after the consumption capacity is considered for the current individuals; omega1The weight of the surplus power index after the consumption capacity is considered for the current individual; rho is a modularity index in the current individual; omega2Is the weight of the modularity index in the current individual.
In practical application, w is increased1The value of (2) can increase the renewable energy consumption capability of the high-voltage power distribution system to be divided and reduce the surplus electric quantity of the system, but the node distribution of the clusters tends to be centralized, the number of the clusters is reduced, and the scale of each cluster is increased; increase w2The value of (a) can lead to better physical structure of the divided clusters, moderate cluster scale and tighter electrical coupling of nodes in the clusters, but the renewable energy consumption capability of the high-voltage power distribution system to be divided can be reduced. The inventor finds that, generally, when aiming at facilitating renewable energy consumption while taking into account the coupling relation of nodes inside the cluster, the following options can be selected: w is a1＝3、w2When the weight combination is 2, the cluster division effect is best.
S104: judging whether the minimum value in the cluster performance indexes respectively corresponding to the current individuals is smaller than a third preset threshold value or whether the iteration times reach a fourth preset threshold value; if yes, go to step S105; if not, executing S106;
specifically, whether the minimum value in the cluster performance indexes respectively corresponding to the current individuals is smaller than a third preset threshold value or not is judged, if yes, the step S105 is executed, and if not, the step S106 is executed.
Specifically, it is determined whether the iteration reaches a fourth preset threshold, if yes, step S105 is executed, and if not, step S106 is executed.
S105: taking the network connection architecture represented by the current individual corresponding to the minimum value of the cluster performance index value as a target connection architecture;
and outputting the cluster division result of the individual representation corresponding to the current iteration as a target connection architecture.
S106: and selecting individuals, crossing the individuals and performing variation processing on the population by using a genetic algorithm, taking the individuals corresponding to the processed network as current individuals, and returning to execute the step S104 until a target connection architecture is obtained.
Specifically, when the individual selection is performed, a selection method according to a conventional genetic algorithm, such as a roulette selection method, may be performed, and the embodiment of the present invention does not limit the selection method of the individual.
Specifically, when individuals are crossed, the adjacency matrix representing the individual i may be arranged in a row in the order of rows, such as: the second row of the individual i adjacent matrix is connected behind the first row of the individual i adjacent matrix, the third row of the individual i adjacent matrix is connected behind the second row of the individual i adjacent matrix, and the like to form a one-dimensional matrix.
The adjacent matrices of the individual i +1 are then arranged in the manner described above to obtain another one-dimensional matrix.
Then, the positions of a plurality of elements at the same position of a row of elements corresponding to the individual i and the individual i +1 are exchanged. The elements that are interchanged include, but are not limited to, the first half, the second half, the middle part, or one or more elements in a row, or a set number of elements in a one-dimensional matrix.
Then, a mutation step is performed, in which mutation operation is performed only on some of the edges where the characteristic values of the connection relationship between the nodes obtained in step S101 are the first preset threshold, that is, "1". The value variation of the operating edge is 1-Va，VaThe edge feature value obtained after the processing in step S102.
It is understood that if the edge between node i and node j has a value of 1, the element in the network or the individual adjacency matrix that characterizes the connection relationship between node i and node j has a value of 1.
After mutation, the process returns to step S104 until the loop ends.
In the prior art, the application of the cluster enables the analysis of the power system to be promoted from a single node to a cluster as a whole. In general, in order to limit the reverse transmission of power to a higher voltage level in a high-voltage distribution network, a method considering renewable energy consumption is often limited to a certain node, for example, a substation and its subordinate area are a node, and only the power and electricity supply and demand balance of the node is taken into consideration during planning, control and operation, but the cooperation between the nodes is ignored, so that the power self-coordination capability in the system cannot be fully exerted, and the maximum renewable energy consumption is realized.
It is to be understood that a prerequisite of the embodiments of the present invention is "inputting the adjacency matrix and node power of the network to be partitioned" shown in fig. 2. In fig. 2, "encoding of genetic algorithm based on adjacency matrix and construction of initial population" corresponds to steps S101 and S102 in the embodiment of the present invention. Steps S1 through S3 in fig. 2 correspond to step S103 of the present invention. The judgment of the iteration end in fig. 2 corresponds to the step S104 in the embodiment of the present invention, and when the judgment result is "Y", the step S105 in the embodiment of the present invention corresponds to the step of outputting the best individual and obtaining the clustering result; if the determination result is "N", the step of selecting the regeneration individual, crossover and mutation corresponds to the step of S106 in the embodiment of the present invention.
By applying the embodiment shown in the figure 1 of the invention, the distributed power supply cluster division is carried out by using the cluster surplus electric quantity index which comprises the electrical distance-based module degree index for community detection and is used for measuring the power consumption capability of the renewable energy source, and the genetic algorithm, compared with the prior art that the cluster division method which simply takes source-load balance as the condition lacks the consideration of the power regulation capability of the system, the method can not ensure that the consumption capability of the system can be fully utilized in the planning and later-period operation of the divided clusters, and the renewable energy consumption capability in the system can be fully exerted, so that the cluster division can be more accurate by applying the embodiment of the invention.
Moreover, by applying the embodiment of the present invention, the structural property of the cluster and the functionality of the cluster can be combined: the defects of the consideration of common structural indexes on the consumption capability of renewable energy sources and the defect of the consideration of indexes for measuring the consumption capability of clusters on the characteristics outside the clusters are overcome. And when the cluster result is obtained, the multi-objective optimization problem is converted into a single objective and given a certain weight coefficient, so that the calculation process is simplified, and the cluster division result achieves a specific balance on the characteristics outside the cluster and the cluster performance.
Moreover, in the embodiment of the invention, the modularity index and the surplus power index both adopt a typical time sequence scene, and are more reasonable compared with cluster division based on a certain time value.
Finally, in the embodiment of the invention, the surplus electric quantity index of the cluster analyzes the renewable energy consumption capability of the cluster from the power regulation perspective, and expands the traditional consideration of cluster source load balance to source load storage balance. Meanwhile, a genetic algorithm coding mode based on a network adjacency matrix is adopted, so that the connectivity of nodes in the cluster is limited, the cluster division meets the requirement of a network grid structure, and the cluster division considering the source network load storage is realized.
Corresponding to the embodiment shown in fig. 1 of the present invention, an embodiment of the present invention further provides a cluster partitioning device for a power distribution network based on high-proportion renewable energy.
Fig. 3 is a schematic structural diagram of a cluster partitioning apparatus for a power distribution network based on high-proportion renewable energy according to an embodiment of the present invention, and as shown in fig. 3, the apparatus includes:
an abstraction module 301, configured to abstract a power distribution network to be subjected to cluster division into a network formed by connecting nodes and edges, represent a connection relationship between two nodes that are connected to each other in the network by using a first preset value, and represent a connection relationship between two nodes that are not connected to each other in the network by using a second preset value, where each power station in the power distribution network is abstracted as a node, and a connection line between the power stations is abstracted as an edge;
A first obtaining module 302, configured to randomly and respectively change a plurality of first preset values in the network to second preset values, obtain a plurality of individuals corresponding to the changed network, and use the individuals as current individuals, where each individual includes at least one cluster, and the cluster is a sub-network formed by connecting a plurality of nodes in the network;
a second obtaining module 303, configured to, for each current individual, obtain a modularity index of the current individual according to a closeness degree of electrical connection between nodes in the current individual, obtain a surplus power index of the current individual according to a power value of each node in the current individual and a power adjustment capability of the current individual for storing energy, and obtain a cluster performance index of the current individual according to the modularity index and the surplus power index;
a determining module 304, configured to determine whether a minimum value in the cluster performance indicators respectively corresponding to the current individuals is smaller than a third preset threshold, or whether the number of iterations reaches a fourth preset threshold;
a first setting module 305, configured to, if the determination result of the determining module 304 is yes, take the connection architecture of the network represented by the current individual corresponding to the minimum value of the cluster performance index value as a target connection architecture;
A second setting module 306, configured to perform individual selection, inter-individual intersection, and individual variation processing on the population by using a genetic algorithm if the determination result of the determining module 304 is negative, and return to the triggering determining module 304 with the individual corresponding to the processed network as the current individual until the target connection architecture is obtained.
By applying the embodiment shown in fig. 3 of the invention, the distributed power supply cluster division is performed by using the cluster surplus power index including the electrical distance-based module degree index for community detection and the index for measuring the power consumption capability of the renewable energy source, and the genetic algorithm, compared with the prior art that the cluster division method simply based on the source-load balance is lack of consideration on the power regulation capability of the system, the divided cluster can not be guaranteed to fully utilize the consumption capability of the system in planning and later-stage operation, and the renewable energy consumption capability in the system is fully exerted, so that the cluster division can be more accurate by applying the embodiment of the invention.
In a specific implementation manner of the embodiment of the present invention, the second obtaining module 303 is configured to:
by means of the formula (I) and (II),
calculating an electrical distance of the current individual, wherein,
by means of the formula (I) and (II),
calculating the electrical distance between node i and node j, wherein,
Lijis the electrical distance between node i and node j; x is the number of the time sections, and t belongs to x; sigma is a summation function;
using the formula, eij＝1-Lij(l), calculating weights of edges between nodes, wherein,
eijThe weight of the edge between the ith node and the jth node; l isijIs the electrical distance between node i and node j; max (l) is the maximum value of the weights in all edges connected to node i;
by means of the formula (I) and (II),
obtaining a modularity index for the current individual, wherein,
rho is a modularity index in the current individual; m is the sum of the weights of all edges in the current individual; e.g. of the typeijThe weight of the edge between the ith node and the jth node; k is a radical ofiIs the sum of the weights of all edges connected to node i; k is a radical ofjIs the sum of the weights of all edges connected to node j; sigma (i, j) is a characteristic parameter of whether the node i and the node j are positioned in the same cluster, and the value of the same cluster is 1 and is different from 0; i is a node serial number; j is the node sequence number.
In a specific implementation manner of the embodiment of the present invention, the second obtaining module 303 is configured to:
by means of the formula (I) and (II),
calculating the net power of the current individual at the time section t with the serial number c, wherein,
by means of the formula (I) and (II), 
Calculating the serial number in the current individual asc power regulation capability of the controllable load of the cluster at time profile t, wherein,
by means of the formula (I) and (II),
calculating the power regulation capacity of the stored energy of the ith energy storage device of the current individual at the moment section t, wherein,
by means of the formula (I) and (II),
calculating the power regulation capacity of the energy storage device of the cluster with the serial number c in the current individual at the moment section t, wherein,
by means of the formula (I) and (II),
calculating the surplus electric quantity index of the cluster with the serial number c in the current individual after power adjustment at the moment section t, wherein,
by means of the formula (I) and (II),
calculating a consideration consumption of a current individualThe surplus power after the capacity index, wherein,
In a specific implementation manner of the embodiment of the present invention, the second obtaining module 303 is configured to:
by means of the formula (I) and (II),
a cluster performance indicator for the current individual is calculated, wherein,
gamma is the cluster performance index of the current individual;
surplus electric quantity indexes after the consumption capacity is considered for the current individuals; omega1The weight of the surplus power index after the consumption capacity is considered for the current individual; rho is a modularity index in the current individual; omega2Is the weight of the modularity index in the current individual.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents and improvements made within the spirit and principle of the present invention are intended to be included within the scope of the present invention.
Claims (6)
1. The cluster division method of the power distribution network based on the high-proportion renewable energy is characterized by comprising the following steps:
1) abstracting a power distribution network to be subjected to cluster division into a network formed by connecting nodes and edges, using a first preset value to represent a connection relation between two nodes which are connected with each other in the network, and using a second preset value to represent a connection relation between two nodes which are not connected with each other in the network, wherein each power station in the power distribution network is abstracted into the nodes, and connecting lines between the power stations are abstracted into the edges;
2) Randomly and respectively changing a plurality of first preset values in the network into second preset values, acquiring a plurality of individuals corresponding to the changed network, and taking the individuals as current individuals, wherein each individual comprises at least one cluster, and the cluster is a sub-network formed by connecting a plurality of nodes in the network;
3) aiming at each current individual, acquiring a modularity index of the current individual according to the electrical connection closeness degree between nodes in the current individual, acquiring a surplus electric quantity index of the current individual according to a power value of each node in the current individual and the power regulation capacity of the current individual for energy storage, and acquiring a cluster performance index of the current individual according to the modularity index and the surplus electric quantity index;
acquiring the surplus power index of the current individual according to the power value of each node in the current individual and the power regulation capacity of the current individual for energy storage, wherein the surplus power index comprises the following steps:
by means of the formula (I) and (II),
calculating the net power of the cluster with the serial number c in the current individual at the moment section t, wherein,
by means of the formula (I) and (II),
calculating the power regulation capacity of the controllable load of the cluster with the serial number c in the current individual at the moment section t, wherein,
by means of the formula (I) and (II),
calculating the power regulation capacity of the stored energy of the ith energy storage device of the current individual at the moment section t, wherein,
by means of the formula (I) and (II),
calculating the power regulation capacity of the energy storage device of the cluster with the serial number c in the current individual at the moment section t, wherein,
by means of the formula (I) and (II),
calculating the surplus electric quantity index of the cluster with the serial number c in the current individual after power adjustment at the moment section t, wherein,
by means of the formula (I) and (II),
and calculating the surplus power index of the current individual after the consumption capacity is considered, wherein,
4) Judging whether the minimum value in the cluster performance indexes respectively corresponding to the current individuals is smaller than a third preset threshold value or whether the iteration times reach a fourth preset threshold value;
5) if so, taking the network connection architecture represented by the current individual corresponding to the minimum value of the cluster performance index value as a target connection architecture;
6) and if not, utilizing a genetic algorithm to perform individual selection, individual crossing and individual variation processing on the population, taking the individual corresponding to the processed network as the current individual, and returning to execute the step 4) until a target connection architecture is obtained.
2. The method for clustering distribution networks based on high-proportion renewable energy according to claim 1, wherein the obtaining the modularity index of the current individual according to the closeness degree of the electrical connection between the nodes in the current individual comprises:
by means of the formula (I) and (II),
calculating an electrical distance of the current individual, wherein,
by means of the formula (I) and (II),
calculating the electrical distance between node i and node j, wherein,
Lijis the electrical distance between node i and node j; x isThe number of the time sections, and t is belonged to x; sigma is a summation function;
using the formula, eij＝1-Lij(l), calculating weights of edges between nodes, wherein,
eijthe weight of the edge between the ith node and the jth node; l isijIs the electrical distance between node i and node j; max (L) is the maximum value of the electrical distance in all sides of the connection;
by means of the formula (I) and (II),
obtaining a modularity index of the current individual, wherein,
Rho is the modularity index of the current individual; m is the sum of the weights of all edges in the current individual; e.g. of the typeijThe weight of the edge between the ith node and the jth node; k is a radical ofiIs the sum of the weights of all edges connected to node i; k is a radical ofjIs the sum of the weights of all edges connected to node j; sigma (i, j) is a characteristic parameter of whether the node i and the node j are positioned in the same cluster, and the value of the same cluster is 1 and is different from 0; i is a node serial number; j is the node sequence number.
3. The method for clustering power distribution networks based on high-proportion renewable energy according to claim 2, wherein the obtaining cluster performance indexes of the current individuals according to the modularity index and the surplus power index comprises:
by means of the formula (I) and (II),
a cluster performance indicator for the current individual is calculated, wherein,
gamma is the cluster performance index of the current individual;
surplus electric quantity indexes after the consumption capacity is considered for the current individuals; omega1Surplus electricity after taking the consumption capacity into consideration for the current individualA weight of the metric; rho is a modularity index in the current individual; omega2Is the weight of the modularity index in the current individual.
4. Cluster partitioning device for a power distribution network based on high proportions of renewable energy, characterized in that said device comprises:
The system comprises an abstraction module, a first storage module and a second storage module, wherein the abstraction module is used for abstracting a power distribution network to be subjected to cluster division into a network formed by connecting nodes and edges, the connection relation between two mutually connected nodes in the network is represented by a first preset value, and the connection relation between two mutually unconnected nodes in the network is represented by a second preset value, each power station in the power distribution network is abstracted into the nodes, and connecting lines between the power stations are abstracted into the edges;
a first obtaining module, configured to randomly and respectively change a plurality of first preset values in the network into second preset values, obtain a plurality of individuals corresponding to the changed network, and use the individuals as current individuals, where each individual includes at least one cluster, and the cluster is a sub-network formed by connecting a plurality of nodes in the network;
a second obtaining module, configured to, for each current individual, obtain a modularity index of the current individual according to a degree of closeness of electrical connection between nodes in the current individual, obtain an excess power index of the current individual according to a power value of each node in the current individual and a power adjustment capability of energy storage of the current individual, and obtain a cluster performance index of the current individual according to the modularity index and the excess power index;
The second obtaining module is configured to:
by means of the formula (I) and (II),
calculating the net power of the cluster with the serial number c in the current individual at the moment section t, wherein,
by means of the formula (I) and (II),
calculating the power regulation capacity of the controllable load of the cluster with the serial number c in the current individual at the moment section t, wherein,
by means of the formula (I) and (II),
calculating the power regulation capacity of the stored energy of the ith energy storage device of the current individual at the moment section t, wherein,
by means of the formula (I) and (II),
calculating the power regulation capacity of the energy storage device of the cluster with the serial number c in the current individual at the moment section t, wherein,
by means of the formula (I) and (II),
calculating the surplus electric quantity index of the cluster with the serial number c in the current individual after power adjustment at the moment section t, wherein,
by means of the formula (I) and (II),
and calculating the surplus power index of the current individual after the consumption capacity is considered, wherein,
the judging module is used for judging whether the minimum value in the cluster performance indexes respectively corresponding to the current individuals is smaller than a third preset threshold value or whether the iteration times reach a fourth preset threshold value;
a first setting module, configured to, if a determination result of the determining module is yes, take a network connection architecture represented by a current individual corresponding to a smallest cluster performance index value in each current individual as a target connection architecture;
and the second setting module is used for selecting individuals, crossing among the individuals and performing variation processing on the individuals on the population by using a genetic algorithm under the condition that the judgment result of the judgment module is negative, and returning the processed individuals corresponding to the network as the current individuals to the triggering judgment module until the target connection architecture is obtained.
5. The apparatus for cluster division of a power distribution network based on high percentage of renewable energy according to claim 4, wherein the second obtaining module is configured to:
By means of the formula (I) and (II),
calculating an electrical distance of the current individual, wherein,
by means of the formula (I) and (II),
calculating the electrical distance between node i and node j, wherein,
Lijis the electrical distance between node i and node j; x is the number of the time sections, and t belongs to x; sigma is a summation function;
Using the formula, eij＝1-Lij(l), calculating weights of edges between nodes, wherein,
eijthe weight of the edge between the ith node and the jth node; l isijIs the electrical distance between node i and node j; max (L) is the maximum value of the electrical distance in all sides of the connection;
by means of the formula (I) and (II),
obtaining a modularity index of the current individual, wherein,
rho is the modularity index of the current individual; m is the sum of the weights of all edges in the current individual; e.g. of the typeijThe weight of the edge between the ith node and the jth node; k is a radical ofiIs the sum of the weights of all edges connected to node i; k is a radical ofjIs the sum of the weights of all edges connected to node j; sigma (i, j) is a characteristic parameter of whether the node i and the node j are positioned in the same cluster, and the value of the same cluster is 1 and is different from 0; i is a node serial number; j is the node sequence number.
6. The apparatus for cluster division of a power distribution network based on high percentage of renewable energy according to claim 5, wherein the second obtaining module is configured to:
by means of the formula (I) and (II),
calculating the currentA cluster performance index for the individual, wherein,
gamma is the cluster performance index of the current individual;
surplus electric quantity indexes after the consumption capacity is considered for the current individuals; omega 1The weight of the surplus power index after the consumption capacity is considered for the current individual; rho is a modularity index in the current individual; omega2Is the weight of the modularity index in the current individual.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811397608.7A CN109829599B (en) | 2018-11-22 | 2018-11-22 | Cluster division method and device for power distribution network based on high-proportion renewable energy |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811397608.7A CN109829599B (en) | 2018-11-22 | 2018-11-22 | Cluster division method and device for power distribution network based on high-proportion renewable energy |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109829599A CN109829599A (en) | 2019-05-31 |
| CN109829599B true CN109829599B (en) | 2022-03-18 |
Family
ID=66859236
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811397608.7A Active CN109829599B (en) | 2018-11-22 | 2018-11-22 | Cluster division method and device for power distribution network based on high-proportion renewable energy |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109829599B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110429649B (en) * | 2019-08-13 | 2021-05-18 | 合肥工业大学 | High-permeability renewable energy cluster division method considering flexibility |
| CN112491057B (en) * | 2020-10-10 | 2022-03-18 | 东北电力大学 | Distributed energy storage control method with aim of eliminating node voltage out-of-limit of power distribution network |
| CN113452053B (en) * | 2021-07-02 | 2023-03-21 | 国网吉林省电力有限公司吉林供电公司 | Distributed energy storage cluster dividing method |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9792647B1 (en) * | 2002-10-08 | 2017-10-17 | Public Service & Gas Company | Electronic bid switch requests for computer-based auctioning of basic generation services |
| CN107732957A (en) * | 2017-08-25 | 2018-02-23 | 合肥工业大学 | A kind of high permeability distribution type renewable energy generating assemblage classification method |
| CN107732945A (en) * | 2017-11-06 | 2018-02-23 | 沈阳工程学院 | A kind of energy-storage units optimization method based on simulated annealing particle cluster algorithm |
| CN108400581A (en) * | 2018-01-31 | 2018-08-14 | 国网浙江省电力公司宁波供电公司 | Isolated island division methods based on energy constraint |
| CN108448620A (en) * | 2018-04-04 | 2018-08-24 | 合肥工业大学 | High permeability distributed generation resource assemblage classification method based on integrated performance index |
-
2018
- 2018-11-22 CN CN201811397608.7A patent/CN109829599B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9792647B1 (en) * | 2002-10-08 | 2017-10-17 | Public Service & Gas Company | Electronic bid switch requests for computer-based auctioning of basic generation services |
| CN107732957A (en) * | 2017-08-25 | 2018-02-23 | 合肥工业大学 | A kind of high permeability distribution type renewable energy generating assemblage classification method |
| CN107732945A (en) * | 2017-11-06 | 2018-02-23 | 沈阳工程学院 | A kind of energy-storage units optimization method based on simulated annealing particle cluster algorithm |
| CN108400581A (en) * | 2018-01-31 | 2018-08-14 | 国网浙江省电力公司宁波供电公司 | Isolated island division methods based on energy constraint |
| CN108448620A (en) * | 2018-04-04 | 2018-08-24 | 合肥工业大学 | High permeability distributed generation resource assemblage classification method based on integrated performance index |
Non-Patent Citations (2)
| Title |
|---|
| Network partition for coordinated control in active distribution networks;BISERICA etal.;《2013 IEEE Grenoble Conference》;20130620;第1-4页 * |
| 含风／光／抽水蓄能并计及负荷响应的海岛微网优化配置;荆朝霞 等;《电力系统自动化》;20170110;第41卷(第1期);第65-72页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109829599A (en) | 2019-05-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109272176B (en) | Calculation method is predicted to platform area line loss per unit using K-means clustering algorithm | |
| CN109599861A (en) | Consider the sending end electric network source structural planning method of local load peak modulation capacity | |
| CN109829599B (en) | Cluster division method and device for power distribution network based on high-proportion renewable energy | |
| CN108448620B (en) | High-permeability distributed power supply cluster division method based on comprehensive performance indexes | |
| CN109390973A (en) | A kind of sending end electric network source structural optimization method considering channel constraint | |
| CN104732300B (en) | A kind of neutral net wind power short term prediction method theoretical based on Fuzzy divide | |
| CN104484833A (en) | Photovoltaic power generation output power tracking algorithm based on genetics algorithm improved RBF-BP neural network | |
| CN109598433A (en) | Consider the sending end electric network source structural planning method of abandoning energy cost and peak regulation demand | |
| CN106355294A (en) | Site selection and capacity determination method of electric vehicle (EV) charging station for large-scale complicated power distribution network | |
| CN106786977B (en) | Charging scheduling method of electric vehicle charging station | |
| CN110119888A (en) | A kind of active gridding planing method based on distributed generation resource access | |
| CN112968441B (en) | Power grid planning method applied to large-scale wind power base | |
| CN111612244B (en) | QRA-LSTM-based method for predicting nonparametric probability of photovoltaic power before day | |
| CN104123684A (en) | Power distribution network economy reconstructing method considering network loss and reliability | |
| CN111092451A (en) | Photovoltaic consumption early warning method based on power distribution network grid | |
| CN111612292A (en) | Cascade hydropower station dispatching control system and method based on key water level control | |
| CN110460038A (en) | It is a kind of to be related to more scene method for expansion planning of power transmission network of new-energy grid-connected | |
| CN115758884A (en) | Digital twin construction method for distributed resources of virtual power plant | |
| CN105930927A (en) | Distribution network planning method based on industry load prediction technology | |
| CN108694475B (en) | Short-time-scale photovoltaic cell power generation capacity prediction method based on hybrid model | |
| CN114417719A (en) | Power distribution network gridding method and device based on graph attention network | |
| CN117114081A (en) | Distributed photovoltaic power prediction method and device based on transfer learning | |
| CN115622056B (en) | Energy storage optimal configuration method and system based on linear weighting and selection method | |
| CN113312779B (en) | High-satisfaction dynamic comprehensive planning method for low-carbon flexible power distribution network | |
| CN109615142A (en) | A kind of wind farm wind velocity combination forecasting method based on wavelet analysis |
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 | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: Tunxi road in Baohe District of Hefei city of Anhui Province, No. 193 230000 Applicant after: Hefei University of Technology Applicant after: STATE GRID ANHUI ELECTRIC POWER Co.,Ltd. Address before: 242000 No. 301, fuming Road, Xuancheng economic and Technological Development Zone, Anhui (HeFei University of Technology Xuancheng campus) Applicant before: Hefei University of Technology Applicant before: STATE GRID ANHUI ELECTRIC POWER Co.,Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |