CN109919801B - Coupling method and device based on node importance of power system - Google Patents

Abstract

The invention discloses a coupling method and a coupling device based on the node importance of a power system, wherein the device is used for realizing the method; implementing cascade failure attack on the power network in the model, dynamically adjusting the actual output power of each power supply node under the condition of ensuring that the power utilization node obtains the conventional input power, obtaining the output power of each power supply node under each stable state in dynamic adjustment, respectively calculating the node influence and the node vulnerability, and obtaining a node d by weighted summation of the node influence and the node vulnerability ₀ The importance of the nodes is ranked from small to small based on the importance of the nodes to generate a new power network, and the new power network is heterogamy coupled with a communication network to generate a new power system model. The invention obtains the node importance through the calculation of the direct current load flow model to obtain a more accurate simulation result, and then performs the different distribution coupling of the new power network and the communication network, thereby improving the robustness of the power system network under the attack.

Description

Coupling method and device based on importance of power system nodes

technical field

The present invention relates to the technical field of network cascading, in particular to a coupling method and device based on the importance of power system nodes.

Background technique

As the most far-reaching technological invention in the new century, the power network has an immeasurable economic and social status. Especially as electricity tends to be shared, socialized, and complicated, the power system presents typical complex characteristics; at the same time, the safety of the power grid, as a key factor affecting the normal life of residents and related to production safety, is more received extensive attention and attention. In recent years, there have been several major power outages in many countries and regions around the world, which not only affected the stable supply of their own power networks, but also brought huge losses to the social economy and production. These blackouts are often caused by a tiny mistake. Therefore, how to avoid or minimize the harm caused by cascading failures in the power system is a hot topic with great practical significance.

As an emerging discipline in the early 21st century, complex network provides a new perspective for us to study and analyze power systems. It maps the entire power system into a power network with a single-layer topology, and through power transmission, Network dynamics ideas such as error propagation and cascading failures explain the potential mechanisms of blackouts and their possible prevention and control measures.

In 2006, a large-scale power outage occurred in the interconnected power grid of the Western European continent. Within ten minutes, a large number of lines and units were tripped one after another, and the entire large power network collapsed into three isolated islands, resulting in a load loss of 17GW and affecting more than 10 million people. The direct and indirect economic losses caused by the normal electricity consumption are immeasurable. The cause of this accident was simply that in order to ensure the safety of the cruise ship, a 380KV double-circuit line on the coast of the western region was disconnected, resulting in a substantial increase in the load in the eastern region and a series of cascading failures. The analysis report of the power company after the event shows that the main problems of the accident are: 1. There is a lack of complete power grid security analysis. The closure of the secondary line can completely avoid the occurrence of this accident. 2. There is a lack of control mechanism for abnormal power flow in the power grid. In the early stage of the accident, only some motors and lines have faults. If the key lines can be cut off in time at this time, the spread of the fault can be cut off and the impact of the accident can be reduced. Therefore, it is of practical significance to analyze and control the power network, that is, to realize the smart grid.

The proposal and development of Cyber-Physical System (CPS) provides new ideas and approaches for promoting the deep integration of power system and information system, and ultimately realizing the goal of grid intelligence. CPS is a new type of system that deeply integrates computer resources with physical systems. By embedding sensor chips in physical entities, it realizes the acquisition and monitoring of physical entity system information; Connect to realize mutual communication and sharing. CPS can also use powerful computer resources to realize factual control and logical management of physical entities, and regularly complete online reorganization and update work. These features make the CPS system highly adaptable, flexible, safe and reliable.

While the application of CPS can bring a lot of convenience to the power system, there are also some drawbacks that need to be solved. Because CPS decomposes the power system into an information network and a power network, the information network implements factual monitoring and management of the power network, and the power network provides the necessary energy to the information network. This also means that the control system is exposed to the environment of the information network, which brings more hidden dangers to the power system. Attacks on the control system may destroy the balance of the power system more quickly and cause large-scale collapse.

In order to improve the security and stability of smart grid, it is necessary to ensure that it still has high robustness when attacked by information network. In the research of physical information fusion system, people often map it as a two-layer interdependent coupling network. The existing research on smart grid mainly includes two aspects: power system model and coupling method between networks. The power system model includes a motherboard model, which mainly considers the topology of the power network, and uses the betweenness of nodes as its load value in the current network; some people propose a load admittance model, which initially sets constant node voltage and edge admittance. , the load value corresponding to each node is calculated by Kirchhoff's law; some people have proposed a DC power flow model, which can better simulate the operation of the power system in reality, and has done a good job in the simulation effect and operation efficiency. a balance. In terms of coupling modes between networks, it mainly includes random coupling, homo-matching coupling and hetero-matching coupling. These three coupling methods connect the nodes in the two networks in a one-to-one correspondence according to the ranking of nodes in a single network. Some people first proposed the concept of two-layer dependent coupled network, and proposed to use the degree or betweenness of nodes as a criterion for sorting; some people proposed that in the power system, the initial load of a node is a parameter that can better reflect the properties of the node.

In the case of a power outage in Western Europe, the node that caused the outage is the western coast node at the edge of the entire power grid. The load of this node during transmission is extremely small, but it causes serious cascading failures.

SUMMARY OF THE INVENTION

The main purpose of the present invention is to propose a coupling method and device based on the importance of power system nodes, aiming at overcoming the above problems.

In order to achieve the above purpose, a coupling method based on the importance of power system nodes proposed by the present invention is characterized in that it includes the following steps:

In S10, the generator node in the power system model is regarded as the power supply node, the consumption node is regarded as the power consumption node, and the electricity transmission line between the nodes is regarded as the connection between the nodes, which is abstracted into the power network; The communication nodes are sorted according to the number of neighbor nodes from small to large to generate a communication network, and a cascading failure model is created by the power network, the communication network and the power supply or power consumption properties of the nodes;

S20 initializes the physical information of all nodes in the cascading failure model, conducts cascading failure attacks on the power network, and dynamically adjusts the actual output power of each power supply node to balance the cascading failure model while ensuring that the power-consuming nodes obtain conventional input power The network load loss ratio is obtained, and the output power of each power supply node in each steady state is obtained in the dynamic adjustment;

S30 uses the DC power flow model to calculate the power value corresponding to each line of the power network according to the output power of each power supply node in each steady state, and obtains the network load sum of the power network in each steady state; it is assumed that there are a total of The number of nodes is n, select any node d ₀ , if the node d ₀ is the attacked node, calculate the network load failure ratio ΔP(d ₀ ) when there is no overload on any line, and regard ΔP(d ₀ ) as the node Influence NI(d ₀ ) of d ₀ ; if node d ₀ is a non-attack node, the survival rate of node d ₀ is calculated according to the state of node d ₀ when other nodes d are attacked, which is regarded as the vulnerability of node d ₀ The weighted summation of the influence NI(d ₀ ) of the node d ₀ and the vulnerability NV(d ₀ ) of the node d ₀ obtains the importance of the node d ₀ , and the n nodes d ₀ are sorted according to their importance from small to large Sort to generate a new power network;

S40 performs heterogeneous coupling DIS between the new power network and the communication network to generate a new power cyber-physical fusion system.

Preferably, the calculation method of the DC power flow model used in the S30 is specifically:

S301 uses the DC power flow model to calculate the load of each line in combination with the power network structure and node attributes, using the following formula:

P _in =T _g P _G -T _d P _D (1)

Θ=BP _in (2)

f _ij =b _ij (θ _i -θ _j ) (3)

和

分别表示供电节点的当前输出功率以及用电节点当前的输入功率，而计算结果P_in是一个n维向量，表示每个节点的功率注入情况，也就是电力来源；In formula (1), it is assumed that the number of nodes in the network is n, the number of power supply nodes is n _g , the number of power consumption nodes is n _d , T _g and T _d are the n*n _g and n*n _d correlation matrices, respectively, When T _g (i, j)=1, it means that the power supply node j corresponds to the network node number i.

and

respectively represent the current output power of the power supply node and the current input power of the power consumption node, and the calculation result P _in is an n-dimensional vector, which represents the power injection of each node, that is, the power source;

In formula (2), the B matrix reflects the power transmission capacity between nodes, that is, the admittance of the line. The calculation result must satisfy Kirchhoff's law, that is, on any node in the circuit, at any time, the flow of The sum of the currents of the nodes is equal to the sum of the currents flowing out of the nodes, and the calculation result Θ=[θ ₁ , θ ₂ ,..., θ _n ] The vector represents the current voltage phase angle of each node;

In formula (3), b _ij is an element in the B matrix, representing the admittance between node i and node j, (θ _i -θ _j ) is the difference between the voltage phase angles between nodes i and j, calculate The result f _ij is the line load between nodes i and j. If f _ij exceeds the rated power, the line will trip due to relay protection, resulting in the generation of new line load, which will lead to a deeper cascade failure;

S302 defines that the influence of a node is expressed as the influence on the entire network after attacking a certain node, and the network load failure ratio ΔP(d ₀ ) when the f _ij of any line is not overloaded is regarded as the influence of the node NI (d ) ₀ ), using the following formula:

表示网络中初始负载之和，P'_D表示当前状态下网络负载之和，而结果ΔP()表示网络中的负载损失比例；where D ₀ represents the initial set of power network nodes, D ₀ -d ₀ represents the set of power network nodes that have cascading failures after the attack,

Represents the sum of the initial load in the network, P' _D represents the sum of the network load in the current state, and the result ΔP() represents the load loss ratio in the network;

S303 defines the vulnerability of a node as the proportion of a node that can survive when other nodes are attacked. The formula is as follows:

Among them, suv(d ₀ , d) represents the state of d ₀ after cascading failure occurs when other nodes d are attacked: survival is 1; otherwise, it is 0, and |D ₀ | represents the number of nodes in the network;

In S304, the weighted summation of the influence NI(d ₀ ) of the node d ₀ and the vulnerability of the node d ₀ obtains the importance of the node d ₀ , and the formula is as follows:

IP(d ₀ )=w·NI(d ₀ )+(1-w)·NV(d ₀ ) (6)

Among them, w is the weight coefficient, which is selected according to the empirical value.

Preferably, the method for obtaining the output power of each power supply node in each steady state in the dynamic adjustment adopts the following formula:

where C _max represents the upper limit of the input power of the power supply node, C _min represents the upper limit of the output power of the power supply node, ΣP _G represents the sum of the output power of the power supply nodes in the current power network, ΣP _D represents the sum of the input power of the power consumption nodes, P( i+1) represents the comparison between the output power of the power supply node in the next steady state and the power in the unstable state, and r represents the ratio at which the generator can adjust its own output.

Preferably, the r value is obtained by performing a cascade failure experiment on a single power network to obtain an r value with moderate experimental effect and consumption.

Preferably, the power system model is a Motter Lai model or an admittance equivalent Admittancemodel model.

Preferably, the S30 can also be completed by using a CFS improved model.

The present invention also discloses a coupling device based on the importance of power system nodes for implementing the above method, which includes:

Create a module to regard the generator node in the power system model as the power supply node, the consumption node as the power consumption node, and the electricity transmission line between the nodes as the connection between the nodes, thus abstracting into a power network; The communication nodes in the system communication network are sorted according to the number of neighbor nodes from small to large to generate the communication network, and the cascade failure model is created by the power network, the communication network and the power supply or power consumption properties of the nodes;

The attack module is used to initialize the physical information of all nodes in the cascading failure model, carry out cascading failure attacks on the power network, and dynamically adjust the actual output power of each power supply node under the condition that the power consumption nodes obtain the regular input power to balance The network load loss ratio of the cascading failure model, and the output power of each power supply node in each steady state is obtained in the dynamic adjustment;

The sorting module is used to calculate the power value corresponding to each line of the power network according to the output power of each power supply node in each steady state, using the DC power flow model, and obtain the network load sum of the power network in each steady state; The total number of nodes in the power network is n, select any node d ₀ , if the node d ₀ is the attacked node, calculate the network load failure ratio ΔP(d ₀ ) when there is no overload on any line, depending on ΔP(d ₀ ) is the influence NI(d ₀ ) of the node d ₀ ; if the node d ₀ is a non-attack node, the survival ratio of the node d ₀ is calculated according to the status of the node d ₀ when other nodes d are attacked, which is regarded as a node The vulnerability of d ₀ ; the weighted summation of the influence NI(d ₀ ) of node d ₀ and the vulnerability of node d ₀ obtains the importance of node d ₀ , and the n nodes d ₀ are sorted according to their importance from small to large generate new electricity networks;

The coupling module is used to perform heterogeneous coupling DIS between the new power network and the communication network to generate a new power cyber-physical fusion system.

Preferably, the sorting module includes:

The DC power flow calculation unit is used to calculate the load condition of each line using the DC power flow model in combination with the power network structure and node attributes, using the following formula:

P _in =T _g P _G -T _d P _D (1)

Θ=BP _in (2)

f _ij =b _ij (θ _i -θ _j ) (3)

和

分别表示供电节点的当前输出功率以及用电节点当前的输入功率，而计算结果P_in是一个n维向量，表示每个节点的功率注入情况，也就是电力来源；In formula (1), it is assumed that the number of nodes in the network is n, the number of power supply nodes is n _g , the number of power consumption nodes is n _d , T _g and T _d are the n*n _g and n*n _d correlation matrices, respectively, When T _g (i, j)=1, it means that the power supply node j corresponds to the network node number i.

and

respectively represent the current output power of the power supply node and the current input power of the power consumption node, and the calculation result P _in is an n-dimensional vector, which represents the power injection of each node, that is, the power source;

In formula (2), the B matrix reflects the power transmission capacity between nodes, that is, the admittance of the line. The calculation result must satisfy Kirchhoff's law, that is, on any node in the circuit, at any time, the flow of The sum of the currents of the nodes is equal to the sum of the currents flowing out of the nodes, and the calculation result Θ=[θ ₁ , θ ₂ ,..., θ _n ] The vector represents the current voltage phase angle of each node;

In formula (3), b _ij is an element in the B matrix, representing the admittance between node i and node j, (θ _i -θ _j ) is the difference between the voltage phase angles between nodes i and j, calculate The result f _ij is the line load between nodes i and j. If f _ij exceeds the rated power, the line will trip due to relay protection, resulting in the generation of new line load, which will lead to a deeper cascade failure;

The influence unit is used to define the influence of a node, which is expressed as the influence on the entire network after attacking a certain node. It is considered that the network load failure ratio ΔP(d ₀ ) when there is no overload in the f _ij of any line is the node’s Influence NI(d ₀ ), using the following formula:

表示网络中初始负载之和，P'_D表示当前状态下网络负载之和，而结果ΔP()表示网络中的负载损失比例；where D ₀ represents the initial set of power network nodes, D ₀ -d ₀ represents the set of power network nodes that have cascading failures after the attack,

Represents the sum of the initial load in the network, P' _D represents the sum of the network load in the current state, and the result ΔP() represents the load loss ratio in the network;

Vulnerability unit, used to define the vulnerability of a node, expressed as the proportion of a node that can survive when other nodes are attacked, the formula is as follows:

Among them, suv(d ₀ , d) represents the state of d ₀ after cascading failure occurs when other nodes d are attacked: survival is 1; otherwise, it is 0, and |D ₀ | represents the number of nodes in the network;

The importance unit is used to obtain the importance of node d ₀ by the weighted summation of the influence NI(d ₀ ) of node d ₀ and the vulnerability of node d ₀ , the formula is as follows:

IP(d ₀ )=w·NI(d ₀ )+(1-w)·NV(d ₀ ) (6)

Among them, w is the weight coefficient, which is selected according to the empirical value.

The invention firstly optimizes the adopted power system model, and proposes a new node importance ranking method by using the DC power flow model calculation. This method separately considers the loss of the entire network load caused by the failure of each node in the power system. degree, that is, the influence of the node, and the survival of the node after the failure of other nodes, that is, the node vulnerability The heterogeneous coupling D method couples a new power network and a communication network, and generates a new power cyber-physical fusion system. This power system network is more robust under attack.

Description of drawings

In order to explain the embodiments of the present invention or the technical solutions in the prior art more clearly, the following briefly introduces the accompanying drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only These are some embodiments of the present invention, and for those of ordinary skill in the art, other drawings can also be obtained according to the structures shown in these drawings without creative efforts.

FIG. 1 is a method flowchart of an embodiment of a coupling method based on the importance of power system nodes of the present invention;

Fig. 2 is the change diagram of the double-layer dependent coupling network of the power cyber-physical system;

The realization, functional characteristics and advantages of the present invention will be further described with reference to the accompanying drawings in conjunction with the embodiments.

Detailed ways

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

It should be noted that if there are directional indications (such as up, down, left, right, front, back, etc.) involved in the embodiments of the present invention, the directional indications are only used to explain a certain posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication also changes accordingly.

In addition, if there are descriptions involving "first", "second", etc. in the embodiments of the present invention, the descriptions of "first", "second", etc. are only used for the purpose of description, and should not be construed as indicating or implying Its relative importance or implicitly indicates the number of technical features indicated. Thus, a feature delimited with "first", "second" may expressly or implicitly include at least one of that feature. In addition, the technical solutions between the various embodiments can be combined with each other, but must be based on the realization by those of ordinary skill in the art. When the combination of technical solutions is contradictory or cannot be realized, it should be considered that the combination of such technical solutions does not exist. , is not within the scope of protection required by the present invention.

As shown in Figure n, a coupling method based on the importance of power system nodes proposed by the present invention includes the following steps:

In S10, the generator node in the power system model is regarded as the power supply node, the consumption node is regarded as the power consumption node, and the electricity transmission line between the nodes is regarded as the connection between the nodes, which is abstracted into the power network; The communication nodes are sorted according to the number of neighbor nodes from small to large to generate a communication network, and a cascading failure model is created by the power network, the communication network and the power supply or power consumption properties of the nodes;

S20 initializes the physical information of all nodes in the cascading failure model, conducts cascading failure attacks on the power network, and dynamically adjusts the actual output power of each power supply node to balance the cascading failure model while ensuring that the power-consuming nodes obtain conventional input power The network load loss ratio is obtained, and the output power of each power supply node in each steady state is obtained in the dynamic adjustment;

S30 uses the DC power flow model to calculate the power value corresponding to each line of the power network according to the output power of each power supply node in each steady state, and obtains the network load sum of the power network in each steady state; it is assumed that there are a total of The number of nodes is n, select any node d ₀ , if the node d ₀ is the attacked node, calculate the network load failure ratio ΔP(d ₀ ) when there is no overload on any line, and regard ΔP(d ₀ ) as the node Influence NI(d ₀ ) of d ₀ ; if node d ₀ is a non-attack node, the survival rate of node d ₀ is calculated according to the state of node d ₀ when other nodes d are attacked, which is regarded as the vulnerability of node d ₀ The weighted summation of the influence NI(d ₀ ) of the node d ₀ and the vulnerability NV(d ₀ ) of the node d ₀ obtains the importance of the node d ₀ , and the n nodes d ₀ are sorted according to their importance from small to large Sort to generate a new power network;

S40 performs heterogeneous coupling DIS between the new power network and the communication network to generate a new power cyber-physical fusion system.

In the embodiment of the present invention, the present invention firstly optimizes the adopted power system model, and proposes a new node importance ranking method by calculating the DC power flow model. After a node failure occurs, the output power of the power supply node and the power consumption node will appear When the input power does not match, dynamically adjust the power of the power supply node and the power consumption node to balance the network load loss ratio of the cascade failure model. The larger the load loss ratio, the better the attack effect, and the greater the degree of network impact. It reflects that this node needs to avoid being attacked more. That is to say, when a network suffers a specific attack, the smaller the proportion of its load loss, the greater the robustness. This method separately considers the loss of the entire network load caused by each node in the power system after its own failure, that is, the influence of the node, and then calculates the survival of the node after other nodes fail, that is, the node vulnerability. The importance of nodes in the network is obtained by the weighted summation of the two, and a new power network is obtained according to the order of node importance. Then, the new power network and communication network are coupled by the heterogeneous coupling DIS method, and the coupling generates a new power cyber-physical system. Converged systems, this power system network is more robust against attacks.

In the embodiment of the present invention, the power system network is a cyber-physical system, and an independent communication network and a power network are coupled to form a two-layer network-dependent network structure. The two-layer network-dependent coupling of the present invention is three steps:

Ranking of communication network nodes: When the communication network is attacked, the attacked node will infect its surrounding nodes with a certain probability and speed, and make other infected nodes infect the fault in the same way. The degree of a node is the number of neighbor nodes of the node. In the network, the nodes with more neighbors, that is, the nodes with higher degrees are more vulnerable to attack and infection, and the present invention also sorts the nodes of the communication network according to the number of neighbor nodes.

Power network node ordering: When the power network is attacked, the fault propagation method is cascading failure, that is, after a node fails, it will cause the node and all the lines connected to it to lose power transmission capability, resulting in more serious damage to the network structure. . The impact of node failure is related to the load and transmission capacity of the node itself. The invention uses a simulation attack model to obtain failure simulation results for the sorting of power network nodes, and then calculates the node importance according to the obtained results, and reorders the nodes according to the node importance to form a new power network.

Coupling of the two networks: Considering that if the nodes in the communication network and the nodes in the power network correspond one-to-one through the coupling edge connection to form a full coupling, in the coupled network, the failure of the node on one side of the dependent edge will directly lead to the other. The nodes on one side are affected, and in the communication network, the node with the higher order is more vulnerable to attack, and the destruction influence of such coupling on the power network is also greater. Therefore, the present invention improves the robustness of the power system network under attack. Robustness, using the heterogeneous coupling (DIS) method, the nodes with low ranking in the power system are coupled and connected with nodes with high ranking in the communication network, and the two networks are reversely coupled and connected in their respective order, thereby maximizing the robustness of the power system network. sex.

Preferably, the calculation method of the DC power flow model used in the S30 is specifically:

S301 uses the DC power flow model to calculate the load of each line in combination with the power network structure and node attributes, using the following formula:

P _in =T _g P _G -T _d P _D (1)

Θ=BP _in (2)

f _ij =b _ij (θ _i -θ _j ) (3)

和

分别表示供电节点的当前输出功率以及用电节点当前的输入功率，而计算结果P_in是一个n维向量，表示每个节点的功率注入情况，也就是电力来源；In formula (1), it is assumed that the number of nodes in the network is n, the number of power supply nodes is n _g , the number of power consumption nodes is n _d , T _g and T _d are the n*n _g and n*n _d correlation matrices, respectively, When T _g (i, j)=1, it means that the power supply node j corresponds to the network node number i.

and

respectively represent the current output power of the power supply node and the current input power of the power consumption node, and the calculation result P _in is an n-dimensional vector, which represents the power injection of each node, that is, the power source;

In formula (2), the B matrix reflects the power transmission capacity between nodes, that is, the admittance of the line. The calculation result must satisfy Kirchhoff's law, that is, on any node in the circuit, at any time, the flow of The sum of the currents of the nodes is equal to the sum of the currents flowing out of the nodes, and the calculation result Θ=[θ ₁ , θ ₂ ,..., θ _n ] The vector represents the current voltage phase angle of each node;

In formula (3), b _ij is an element in the B matrix, representing the admittance between node i and node j, (θ _i -θ _j ) is the difference between the voltage phase angles between nodes i and j, calculate The result f _ij is the line load between nodes i and j. If f _ij exceeds the rated power, the line will trip due to relay protection, resulting in the generation of new line load, which will lead to a deeper cascade failure;

S302 defines that the influence of a node is expressed as the influence on the entire network after attacking a certain node, and the network load failure ratio ΔP(d ₀ ) when the f _ij of any line is not overloaded is regarded as the influence of the node NI (d ) ₀ ), using the following formula:

表示网络中初始负载之和，P'_D表示当前状态下网络负载之和，而结果ΔP表示网络中的负载损失比例；where D ₀ represents the initial set of power network nodes, D ₀ -d ₀ represents the set of power network nodes that have cascading failures after the attack,

represents the sum of the initial load in the network, P' _D represents the sum of the network load in the current state, and the result ΔP represents the load loss ratio in the network;

S303 defines the vulnerability of a node as the proportion of a node that can survive when other nodes are attacked. The formula is as follows:

Among them, suv(d ₀ , d) represents the state of d ₀ after cascading failure occurs when other nodes d are attacked: survival is 1; otherwise, it is 0, and |D ₀ | represents the number of nodes in the network;

In S304, the weighted summation of the influence NI(d ₀ ) of the node d ₀ and the vulnerability of the node d ₀ obtains the importance of the node d ₀ , and the formula is as follows:

IP(d ₀ )=w·NI(d ₀ )+(1-w)·NV(d ₀ ) (6)

Among them, w is the weight coefficient, which is selected according to the empirical value.

Preferably, the method for obtaining the output power of each power supply node in each steady state in the dynamic adjustment adopts the following formula:

where C _max represents the upper limit of the input power of the power supply node, C _min represents the upper limit of the output power of the power supply node, ΣP _G represents the sum of the output power of the power supply nodes in the current power network, ΣP _D represents the sum of the input power of the power consumption nodes, P( i+1) represents the comparison between the output power of the power supply node in the next steady state and the power in the unstable state, and r represents the ratio at which the generator can adjust its own output.

Preferably, the r value is obtained by performing a cascade failure experiment on a single power network to obtain an r value with moderate experimental effect and consumption.

In the embodiment of the present invention, considering that the actual output of the power supply node is also affected by its own rated power, under theoretical conditions, the value of r should be as large as possible, but excessively large adjustment power will require higher generator physical performance, resulting in additional consumption. In order to optimize the physical performance of the generator, a cascade failure experiment can be performed on the r value in the formula in a single power network, and the r value with moderate effect and consumption can be selected.

Preferably, the power system model is a Motter Lai model or an admittance equivalent Admittancemodel model.

Preferably, the S30 can also be completed by using a CFS improved model.

The present invention also discloses a coupling device based on the importance of power system nodes for implementing the above method, which includes:

Create a module to regard the generator node in the power system model as the power supply node, the consumption node as the power consumption node, and the electricity transmission line between the nodes as the connection between the nodes, thus abstracting into a power network; The communication nodes in the system communication network are sorted according to the number of neighbor nodes from small to large to generate the communication network, and the cascade failure model is created by the power network, the communication network and the power supply or power consumption properties of the nodes;

The attack module is used to initialize the physical information of all nodes in the cascading failure model, carry out cascading failure attacks on the power network, and dynamically adjust the actual output power of each power supply node under the condition that the power consumption nodes obtain the regular input power to balance The network load loss ratio of the cascading failure model, and the output power of each power supply node in each steady state is obtained in the dynamic adjustment;

The sorting module is used to calculate the power value corresponding to each line of the power network according to the output power of each power supply node in each steady state, using the DC power flow model, and obtain the network load sum of the power network in each steady state; The total number of nodes in the power network is n, select any node d ₀ , if the node d ₀ is the attacked node, calculate the network load failure ratio ΔP(d ₀ ) when there is no overload on any line, depending on ΔP(d ₀ ) is the influence NI(d ₀ ) of the node d ₀ ; if the node d ₀ is a non-attack node, the survival ratio of the node d ₀ is calculated according to the status of the node d ₀ when other nodes d are attacked, which is regarded as a node The vulnerability of d ₀ ; the weighted summation of the influence NI(d ₀ ) of node d ₀ and the vulnerability of node d ₀ obtains the importance of node d ₀ , and the n nodes d ₀ are sorted according to their importance from small to large generate new electricity networks;

The coupling module is used to perform heterogeneous coupling DIS between the new power network and the communication network to generate a new power cyber-physical fusion system.

Preferably, the sorting module includes:

The DC power flow calculation unit is used to calculate the load condition of each line using the DC power flow model in combination with the power network structure and node attributes, using the following formula:

P _in =T _g P _G -T _d P _D (1)

Θ=BP _in (2)

f _ij =b _ij (θ _i -θ _j ) (3)

和

分别表示供电节点的当前输出功率以及用电节点当前的输入功率，而计算结果P_in是一个n维向量，表示每个节点的功率注入情况，也就是电力来源；In formula (1), it is assumed that the number of nodes in the network is n, the number of power supply nodes is n _g , the number of power consumption nodes is n _d , T _g and T _d are the n*n _g and n*n _d correlation matrices, respectively, When T _g (i, j)=1, it means that the power supply node j corresponds to the network node number i.

and

respectively represent the current output power of the power supply node and the current input power of the power consumption node, and the calculation result P _in is an n-dimensional vector, which represents the power injection of each node, that is, the power source;

In formula (2), the B matrix reflects the power transmission capacity between nodes, that is, the admittance of the line. The calculation result must satisfy Kirchhoff's law, that is, on any node in the circuit, at any time, the flow of The sum of the currents of the nodes is equal to the sum of the currents flowing out of the nodes, and the calculation result Θ=[θ ₁ , θ ₂ ,..., θ _n ] The vector represents the current voltage phase angle of each node;

In formula (3), b _ij is an element in the B matrix, representing the admittance between node i and node j, (θ _i -θ _j ) is the difference between the voltage phase angles between nodes i and j, calculate The result f _ij is the line load between nodes i and j. If f _ij exceeds the rated power, the line will trip due to relay protection, resulting in the generation of new line load, which will lead to a deeper cascade failure;

The influence unit is used to define the influence of a node, which is expressed as the influence on the entire network after attacking a certain node. It is considered that the network load failure ratio ΔP(d ₀ ) when there is no overload in the f _ij of any line is the node’s Influence NI(d ₀ ), using the following formula:

表示网络中初始负载之和，P'_D表示当前状态下网络负载之和，而结果ΔP()表示网络中的负载损失比例；where D ₀ represents the initial set of power network nodes, D ₀ -d ₀ represents the set of power network nodes that have cascading failures after the attack,

Represents the sum of the initial load in the network, P' _D represents the sum of the network load in the current state, and the result ΔP() represents the load loss ratio in the network;

Vulnerability unit, used to define the vulnerability of a node, expressed as the proportion of a node that can survive when other nodes are attacked, the formula is as follows:

Among them, suv(d ₀ , d) represents the state of d ₀ after cascading failure occurs when other nodes d are attacked: survival is 1; otherwise, it is 0, and |D ₀ | represents the number of nodes in the network;

The importance unit is used to obtain the importance of node d ₀ by the weighted summation of the influence NI(d ₀ ) of node d ₀ and the vulnerability of node d ₀ , the formula is as follows:

IP(d ₀ )=w·NI(d ₀ )+(1-w)·NV(d ₀ ) (6)

Among them, w is the weight coefficient, which is selected according to the empirical value.

Since the device adopts all the technical solutions of all the embodiments of the above methods, it has at least all the beneficial effects brought about by the technical solutions of the above-mentioned embodiments, which will not be repeated here.

The above descriptions are only the preferred embodiments of the present invention, and are not intended to limit the scope of the present invention. Under the inventive concept of the present invention, the equivalent structural transformations made by the contents of the description and drawings of the present invention, or the direct/indirect application Other related technical fields are included in the scope of patent protection of the present invention.

Claims (6)

1. A coupling method based on node importance of a power system is characterized by comprising the following steps:

s10, regarding the generator nodes in the power system model as power supply nodes, regarding the consumption nodes as power utilization nodes, regarding the electrical transmission lines between the nodes as connecting lines between the nodes, and abstracting the power system model into a power network; the method comprises the steps that communication nodes in a communication network of the power system are sequenced from small to large according to the number of neighbor nodes to generate the communication network, and a cascade failure model is established according to power supply or power utilization attributes of the power network, the communication network and the nodes;

s20, initializing physical information of all nodes in the cascade failure model, carrying out cascade failure attack on the power network, dynamically adjusting the actual output power of each power supply node under the condition that the power utilization node obtains conventional input power, and obtaining the output power of each power supply node under each stable state in dynamic adjustment;

s30, according to the output power of each power supply node in each stable state, calculating a power value corresponding to each line of the power network by using a direct current load flow model, and acquiring the network load sum of the power network in each stable state; assuming that the number of common nodes in the power network is n, any node d is selected ₀ If node d ₀ Calculating the network load failure proportion delta P (d) when any line has no overload for the attacked node ₀ ) Apparent Δ P (d) ₀ ) Is node d ₀ Influence of (d) NI (d) ₀ ) (ii) a If node d ₀ If the node is a non-attack node, the node d is attacked according to other nodes ₀ State situation calculation node d ₀ Is considered as node d ₀ (iii) vulnerability of; to node d ₀ Influence of (d) NI (d) ₀ ) And node d ₀ Vulnerability NV (d) of ₀ ) Weighted summation obtaining node d ₀ Importance of, n nodes d ₀ Generating new power networks in a descending order according to the importance of the power networks;

s40, carrying out different distribution coupling DIS on the new power network and the communication network to generate a new power information physical fusion system;

the calculation method using the direct current power flow model in the step S30 specifically includes:

s301, combining the structure and the node attribute of the power network, calculating the load condition of each line by using a direct current load flow model, and adopting the following formula:

P _in ＝T _g P _G -T _d P _D (1)

Θ＝BP _in (2)

f _ij ＝b _ij (θ _i -θ _j ) (3)

in formula (1), it is assumed that the number of nodes in the network is n, and the number of power supply nodes is n _g The number of the power nodes is n _d ，T _g And T _d Are each n x n _g And n x n _d Of (2), wherein T _g When (i, j) is 1, the power supply node j corresponds to the network node number i;

and

respectively representing the current output power of the power supply node and the current input power of the power utilization node, and calculating the result P _in Is an n-dimensional vector representing the power injection condition of each node, namely the power source;

in the formula (2), the B matrix reflects the power transmission capability between nodes, that is, the admittance of the line, and the calculation result must satisfy kirchhoff's law, that is, at any node in the circuit, the sum of the currents flowing into the nodes is equal to the sum of the currents flowing out of the nodes at any time, and the calculation result Θ is [ θ ═ θ [ ] ₁ ,θ ₂ ,…,θ _n ]The vector represents the current voltage phase angle of each node;

in the formula (3), b _ij Is an element in the B matrix representing the admittance between node i and node j, (θ) _i -θ _j ) The result f is calculated as the difference in voltage phase angle between the i and j nodes _ij That is the line load between nodes i and j, if f _ij Beyond the rated power, the line trips for relay protection, resulting in the production of new line loadsThus, a deeper level of cascade failure can be caused;

s302, defining the influence of the node as the influence of attacking a certain node on the whole network, and considering f of any line _ij Network load failure proportion delta P (d) when no load overload exists ₀ ) Is the influence of the node NI (d) ₀ ) The following formula is adopted:

wherein D ₀ Representing an initial set of power network nodes, D ₀ -d ₀ A collection of power network nodes representing a cascade failure after an attack,

denotes the sum of the initial loads in the network, P' _D Represents the sum of the network loads in the current state, and the result is Δ P (d) ₀ ) Representing the proportion of load loss in the network;

s303 defines the vulnerability of a node as the proportion of a certain node that can survive when other nodes are attacked, and the formula is as follows:

wherein suv (d) ₀ D) indicates that d) after cascade failure occurs when other nodes d are attacked ₀ The state of (2): survival was 1; otherwise 0, | D ₀ L represents the number of nodes in the network;

s304 pairs of nodes d ₀ Influence of (d) NI (d) ₀ ) And node d ₀ Obtaining node d by weighted summation of vulnerabilities ₀ The formula is as follows:

IP(d ₀ )＝w·NI(d ₀ )+(1-w)·NV(d ₀ ) (6)

wherein w is a weight coefficient and is taken according to an empirical value.

2. The power system node importance-based coupling method according to claim 1, wherein the method for obtaining the output power of each power supply node in each steady state in dynamic regulation adopts the following equation:

wherein C is _max Representing the upper limit of the input power of the supply node, C _min Representing the upper output power limit of the supply node, sigma P _G Representing the sum of the output powers, Σ P, of the power supply nodes in the current power network _D Represents the sum of the input power of the power utilization nodes, P (i +1) represents the output power of the power supply node in the next steady state and the power in the unstable state, and r represents the ratio of the self-output of the generator which can be adjusted.

3. The coupling method based on the importance of the nodes of the power system as claimed in claim 2, wherein the r value is obtained by performing a cascading failure experiment on a single power network, and the r value is moderate in experimental effect and consumption.

4. The coupling method based on the importance of the power system nodes as claimed in claim 1, wherein the power system model is a molelay Motter Lai model or an Admittance equivalent acceptance model.

5. The coupling method based on node importance of power system of claim 1, wherein the S30 is further performed using a CFS improved model.

6. A coupling device based on power system node importance, comprising:

the system comprises a creation module, a power supply module, a power consumption module and a power grid module, wherein the creation module is used for regarding a generator node in a power system model as a power supply node, regarding a consumption node as a power utilization node and regarding an electrical transmission line between nodes as a connection line between the nodes, so that the nodes are abstracted into a power grid; the method comprises the steps that communication nodes in a communication network of the power system are sequenced from small to large according to the number of neighbor nodes to generate the communication network, and a cascade failure model is established according to power supply or power utilization attributes of the power network, the communication network and the nodes;

the attack module is used for initializing physical information of all nodes in the cascade failure model, carrying out cascade failure attack on the power network, dynamically adjusting the actual output power of each power supply node under the condition that the power utilization node obtains the conventional input power, and obtaining the output power of each power supply node under each stable state in the dynamic adjustment;

the sorting module is used for calculating a power value corresponding to each line of the power network by using a direct current load flow model according to the output power of each power supply node in each stable state, and acquiring the sum of network loads of the power network in each stable state; assuming that the number of common nodes in the power network is n, any node d is selected ₀ If node d ₀ Calculating the network load failure proportion delta P (d) when any line has no overload for the attacked node ₀ ) Apparent Δ P (d) ₀ ) Is node d ₀ Influence of (d) NI (d) ₀ ) (ii) a If node d ₀ If the node is a non-attack node, the node d is attacked according to other nodes ₀ State situation calculation node d ₀ Is considered as node d ₀ (iii) vulnerability of; to node d ₀ Influence of (d) NI (d) ₀ ) And node d ₀ Obtaining node d by weighted summation of vulnerabilities ₀ Importance of, n nodes d ₀ Generating new power networks in a descending order according to the importance of the power networks;

the coupling module is used for carrying out different distribution coupling DIS on the new power network and the communication network to generate a new power information physical fusion system;

the sorting module comprises:

the direct current load flow calculation unit is used for calculating the load condition of each line by using a direct current load flow model in combination with the power network structure and the node attributes, and adopts the following formula:

P _in ＝T _g P _G -T _d P _D (1)

Θ＝BP _in (2)

f _ij ＝b _ij (θ _i -θ _j ) (3)

in formula (1), it is assumed that the number of nodes in the network is n, and the number of power supply nodes is n _g The number of the power nodes is n _d ，T _g And T _d Are each n x n _g And n x n _d Wherein T is _g When (i, j) is 1, the power supply node j corresponds to the network node number i;

and

respectively representing the current output power of the power supply node and the current input power of the power utilization node, and calculating the result P _in Is an n-dimensional vector representing the power injection condition of each node, namely the power source;

in the formula (2), the B matrix reflects the power transmission capability between nodes, that is, the admittance of the line, and the calculation result must satisfy kirchhoff's law, that is, at any node in the circuit, the sum of the currents flowing into the nodes is equal to the sum of the currents flowing out of the nodes at any time, and the calculation result Θ is [ θ ═ θ [ ] ₁ ,θ ₂ ,…,θ _n ]The vector represents the current voltage phase angle of each node;

in the formula (3), b _ij Is an element in the B matrix representing the admittance between node i and node j, (θ) _i -θ _j ) The result f is calculated as the difference in voltage phase angle between the i and j nodes _ij Is the line load between nodes i and j, if f _ij When the rated power is exceeded, the circuit trips due to relay protection, so that new circuit load is generated, and further cascade failure is caused;

an influence unit for defining the influence of the node as the influence of the node on the whole network, and considering any oneF of the line _ij Network load failure proportion delta P (d) when no load overload exists ₀ ) As influence of the node NI (d) ₀ ) The following formula is adopted:

wherein D ₀ Representing an initial set of power network nodes, D ₀ -d ₀ A collection of power network nodes representing a cascade failure after an attack,

representing the sum of initial loads in the network, P' _D Represents the sum of the network loads in the current state, and the result is Δ P (d) ₀ ) Representing the proportion of load loss in the network;

the vulnerability unit is used for defining the vulnerability of the nodes to be expressed as the proportion of the nodes which can survive when other nodes are attacked, and the formula is as follows:

wherein suv (d) ₀ D) indicates that d) after cascade failure occurs when other nodes d are attacked ₀ The state of (2): survival was 1; otherwise 0, | D ₀ L represents the number of nodes in the network;

significance unit for pair of nodes d ₀ Influence of (d) NI (d) ₀ ) And node d ₀ Obtaining node d by weighted summation of vulnerabilities ₀ The formula is as follows:

IP(d ₀ )＝w·NI(d ₀ )+(1-w)·NV(d ₀ ) (6)

wherein w is a weight coefficient and is taken according to an empirical value.

Images