阅读说明：本技术 一种考虑网络攻击的电力系统预想故障集生成方法和装置 (Power system expected fault set generation method and device considering network attack ) 是由 倪明 童和钦 李满礼 赵丽莉 张迎星 王�琦 蔡星浦 于 2020-03-10 设计创作，主要内容包括：本发明公开了一种考虑网络攻击的电力系统预想故障集生成方法和装置,根据设定的网络攻击类型和攻击线路数量确定网络攻击线路的成功率,通过对某个或某些线路进行网络攻击成功的状态和攻击成功后电力系统中其他线路处于正常或断线状态的排列组合构建电力系统攻击成功状态空间；对于配电网,遍历电力系统攻击成功状态空间,对于输电网,对电力系统攻击成功状态空间先部分遍历再抽样遍历,分别计算网络攻击成功线路断线导致电力系统损失的负荷量,基于攻击类型、攻击成功的线路和攻击成功线路断线导致电力系统损失的非零负荷量确定考虑网络攻击的预想故障集。本发明得到考虑网络攻击的电力系统预想故障集,对现有三道防线中的预想故障集进行补充。(The invention discloses a method and a device for generating an expected fault set of a power system considering network attack, wherein the success rate of network attack lines is determined according to the set network attack type and the attack line number, and a power system attack success state space is constructed by the successful network attack state of a certain line or some lines and the arrangement and combination of the normal or disconnection state of other lines in the power system after the successful attack; for a power distribution network, traversing the power system attack success state space, for a power transmission network, traversing the power system attack success state space partially and then sampling, respectively calculating the load quantity of the power system loss caused by the network attack success line disconnection, and determining an expected fault set considering the network attack based on the attack type, the line successful in the attack and the non-zero load quantity of the power system loss caused by the attack success line disconnection. The invention obtains the expected failure set of the power system considering the network attack and supplements the expected failure set in the existing three defense lines.)

1. A power system expected fault set generation method considering network attacks is characterized by comprising the following steps: the method comprises the following steps:

determining the success rate of the network attack lines according to the set network attack type and the attack line number, and constructing a power system attack success state space by the permutation and combination of the successful state of the network attack on a certain line or some lines and the normal or disconnection state of other lines in the power system after the attack is successful;

for the power distribution network, traversing a power system attack success state space, respectively calculating the load quantity of the power system loss caused by the disconnection of the network attack success line, and determining an expected fault set of the network attack based on the attack type, the attack success line and the non-zero load quantity of the power system loss caused by the disconnection of the attack success line;

for the power transmission network, the state space of the power system attack success state is firstly partially traversed and then is sampled and traversed, the load quantity of the power system loss caused by the network attack success line disconnection is respectively calculated, and the expected fault set of the network attack is determined based on the attack type, the line of the attack success and the non-zero load quantity of the power system loss caused by the attack success line disconnection.

2. The method for generating the expected failure set of the power system considering the cyber attack as claimed in claim 1, wherein: the load quantity of the power system loss caused by the line disconnection of the network attack success adopts an optimal load reduction calculation method based on a direct current load flow model, and the optimization objective of the calculation method is as follows:

minf＝∑_t∈TL_t

in the formula: f represents the sum of the load shedding amount of the power system; t represents a power system node set; t represents the node orderNumber L_tRepresenting the load shedding value of the node t;

the constraint conditions include:

in the formula: w represents a power system line set; l represents a line serial number; f_lRepresenting the power flow on line l; z_lRepresenting the state of the line l, whether the line is normally connected or disconnected; x is the number of_lRepresenting the impedance of the line l, H_tA correlation matrix representing the nodes t is formed,_trepresenting a node t phase angle matrix;

∑_m∈MB_mG_m-∑_l∈WH_tF_l＝Q_t-L_tt∈T

in the formula: m represents a system generator set, M represents a generator serial number, B_mRepresenting the state of the generator, normal operation or fault not-operating, G_mRepresenting the generated power of generator m, Q_tRepresenting the load on node t;

-F_l ^max≤F_l≤F_l ^maxl∈W

in the formula, F_l ^maxRepresenting the transmission limit of the line l, G_m ^minAnd G_m ^maxRepresenting the maximum and minimum output of the generator m, respectively.

3. The method for generating the expected failure set of the power system considering the cyber attack as claimed in claim 1, wherein: for the power transmission network, the state space of the power system attack success is firstly traversed partially and then is traversed by sampling, and the method comprises the following steps:

multiplying the probabilities of the line states of all permutation and combination in the power system attack success state space respectively to obtain the occurrence probability of all the states in the state space, and sequencing the occurrence probability;

traversing the line states from front to back according to the sequencing result according to the set number of the states needing to be traversed;

and randomly sampling the rest line states according to the set number of the states needing to be sampled.

4. The method for generating the expected failure set of the power system considering the cyber attack as claimed in claim 3, wherein: the sampling method adopts a Monte Carlo method.

5. The method for generating the expected failure set of the power system considering the cyber attack as claimed in claim 1, wherein: the network attack types include spurious data injection attacks against power system measurements, replay attacks against power system instructions, threshold modification attacks against power system protection devices.

6. An expected failure set generation device for an electric power system considering network attacks, comprising: the method comprises the following steps:

the power system attack success state space construction module is used for determining the success rate of the network attack lines according to the set network attack type and the attack line number, and constructing the power system attack success state space by carrying out the arrangement combination of the network attack success state on a certain line or some lines and the normal or disconnection state of other lines in the power system after the attack success;

the expected fault set generation module considering the network attack is used for traversing the attack success state space of the power system for the power distribution network, respectively calculating the load quantity of the power system loss caused by the disconnection of the successful line of the network attack, and determining an expected fault set considering the network attack based on the attack type, the line successful in the attack and the non-zero load quantity of the power system loss caused by the disconnection of the successful line of the attack;

for the power transmission network, the state space of the power system attack success state is firstly partially traversed and then is sampled and traversed, the load quantity of the power system loss caused by the network attack success line disconnection is respectively calculated, and an expected fault set considering the network attack is determined based on the attack type, the line of the attack success and the non-zero load quantity of the power system loss caused by the attack success line disconnection.

7. The expected failure set generation device for power system considering network attack according to claim 6, wherein: the load quantity of the power system loss caused by the line disconnection of the network attack success adopts an optimal load reduction calculation method based on a direct current load flow model, and the optimization objective of the calculation method is as follows:

minf＝∑_t∈TL_t

wherein f represents the sum of the load shedding amount of the power system, T represents the node set of the power system, T represents the node serial number, L_tRepresenting the load shedding value of the node t;

the constraint conditions include:

in the formula: w represents a power system line set; l represents a line serial number; f_lRepresenting the power flow on line l; z_lRepresenting the state of the line l, whether the line is normally connected or disconnected; x is the number of_lRepresenting the impedance of the line l, H_tA correlation matrix representing the nodes t is formed,_trepresenting a node t phase angle matrix;

∑_m∈MB_mG_m-∑_l∈WH_tF_l＝Q_t-L_tt∈T

in the formula: m represents a system generator set, M represents a generator serial number, B_mRepresenting the state of the generator, normal operation or fault not-operating, G_mRepresenting the generated power of generator m, Q_tRepresenting the load on node t;

-F_l ^max≤F_l≤F_l ^maxl∈W

in the formula, F_l ^maxRepresenting the transmission limit of the line l, G_m ^minAnd G_m ^maxRepresenting the maximum and minimum output of the generator m, respectively.

8. The expected failure set generation device for power system considering network attack according to claim 6, wherein: for the power transmission network, the state space of the power system attack success is firstly traversed partially and then is traversed by sampling, and the method comprises the following steps:

multiplying the probabilities of the line states of all permutation and combination in the power system attack success state space respectively to obtain the occurrence probability of all the states in the state space, and sequencing the occurrence probability;

traversing the line states from front to back according to the sequencing result according to the set number of the states needing to be traversed;

and randomly sampling the rest line states according to the set number of the states needing to be sampled.

9. The expected failure set generation device for power system considering cyber attack according to claim 8, wherein: the sampling method adopts a Monte Carlo method.

10. The expected failure set generation device for power system considering network attack according to claim 6, wherein: the network attack types include spurious data injection attacks against power system measurements, replay attacks against power system instructions, threshold modification attacks against power system protection devices.

Technical Field

The invention relates to the technical field of power system network security, in particular to a power system expected fault set generation method and device considering network attack.

Background

As one of the most important infrastructures in modern society, power systems are gradually merged with communication systems, and become typical Cyber Physical Systems (CPS). On one hand, the deep coupling of information physics enables the power grid to become more intelligent; on the other hand, it also carries a potential risk of cyber attack. The conventional power system expected fault set only considers faults caused by natural factors, and does not consider power system faults caused by network attacks. When a fault may occur, the dispatcher cannot match a proper expected fault, and the subsequent control effect is further influenced. Therefore, it is necessary to generate a targeted set of expected failures in view of the possibility of a network attack.

Disclosure of Invention

In order to solve the defects in the prior art, the invention provides a power system expected fault set generation method and device considering network attacks, and solves the problem that the network attacks are not considered in the current power system expected faults.

In order to achieve the above purpose, the invention adopts the following technical scheme: a power system expected fault set generation method considering network attacks comprises the following steps:

determining the success rate of the network attack lines according to the set network attack type and the attack line number, and constructing a power system attack success state space by the permutation and combination of the successful state of the network attack on a certain line or some lines and the normal or disconnection state of other lines in the power system after the attack is successful;

for the power distribution network, traversing a power system attack success state space, respectively calculating the load quantity of the power system loss caused by the disconnection of the network attack success line, and determining an expected fault set considering the network attack based on the attack type, the line with successful attack and the non-zero load quantity of the power system loss caused by the disconnection of the attack success line;

for the power transmission network, the state space of the power system attack success state is firstly partially traversed and then is sampled and traversed, the load quantity of the power system loss caused by the network attack success line disconnection is respectively calculated, and an expected fault set considering the network attack is determined based on the attack type, the line of the attack success and the non-zero load quantity of the power system loss caused by the attack success line disconnection.

Further, the load quantity of the power system loss caused by the line disconnection of the network attack success adopts an optimal load reduction calculation method based on a direct current load flow model, and the optimization objective of the calculation method is as follows:

the power grid is attacked by the network, so that the loss load is minimum when the topology is changed:

minf＝∑_t∈TL_t

wherein f represents the sum of the load shedding amount of the power system, T represents the node set of the power system, T represents the node serial number, L_tRepresenting the load shedding value of the node t;

the constraint conditions include:

in the formula: w represents a power system line set; l represents a line serial number; f_lRepresenting the power flow on line l; z_lRepresenting the state of the line l, whether the line is normally connected or disconnected; x is the number of_lRepresenting the impedance of the line l, H_tA correlation matrix representing the nodes t is formed,_trepresenting a node t phase angle matrix;

∑_m∈MB_mG_m-∑_l∈WH_tF_l＝Q_t-L_tt∈T

in the formula: m represents a system generator set, M represents a generator serial number, B_mRepresenting the state of the generator, normal operation or fault not-operating, G_mRepresenting the generated power of generator m, Q_tRepresenting the load on node t;

-F_l ^max≤F_l≤F_l ^maxl∈W

in the formula, F_l ^maxRepresenting the transmission limit of the line l, G_m ^minAnd G_m ^maxRepresenting the maximum and minimum output of the generator m, respectively.

Further, for the power transmission network, the state space of the power system attack success is firstly traversed partially and then traversed by sampling, and the method comprises the following steps:

multiplying the probabilities of the line states of all permutation and combination in the power system attack success state space respectively to obtain the occurrence probability of all the states in the state space, and sequencing the occurrence probability;

traversing the line states from front to back according to the sequencing result according to the set number of the states needing to be traversed;

and randomly sampling the rest line states according to the set number of the states needing to be sampled.

Further, the sampling method adopts a Monte Carlo method.

Further, the network attack types include a spurious data injection attack against power system measurements, a replay attack against power system instructions, and a threshold modification attack against power system protection devices.

An expected failure set generation apparatus for an electric power system considering network attacks, comprising:

the power system attack success state space construction module is used for determining the success rate of the network attack lines according to the set network attack type and the attack line number, and constructing the power system attack success state space by carrying out the arrangement combination of the network attack success state on a certain line or some lines and the normal or disconnection state of other lines in the power system after the attack success;

the expected fault set generation module considering the network attack is used for traversing the attack success state space of the power system for the power distribution network, respectively calculating the load quantity of the power system loss caused by the disconnection of the successful line of the network attack, and determining an expected fault set considering the network attack based on the attack type, the line successful in the attack and the non-zero load quantity of the power system loss caused by the disconnection of the successful line of the attack;

for the power transmission network, the state space of the power system attack success state is firstly partially traversed and then is sampled and traversed, the load quantity of the power system loss caused by the network attack success line disconnection is respectively calculated, and an expected fault set considering the network attack is determined based on the attack type, the line of the attack success and the non-zero load quantity of the power system loss caused by the attack success line disconnection.

Further, the load quantity of the power system loss caused by the line disconnection of the network attack success adopts an optimal load reduction calculation method based on a direct current load flow model, and the optimization objective of the calculation method is as follows:

minf＝∑_t∈TL_t

in the formula: f represents the sum of the load shedding amount of the power system; t represents a power system node set; t represents the node orderNumber L_tRepresenting the load shedding value of the node t;

the constraint conditions include:

in the formula: w represents a power system line set; l represents a line serial number; f_lRepresenting the power flow on line l; z_lRepresenting the state of the line l, whether the line is normally connected or disconnected; x is the number of_lRepresenting the impedance of the line l, H_tA correlation matrix representing the nodes t is formed,_trepresenting a node t phase angle matrix;

∑_m∈MB_mG_m-∑_l∈WH_tF_l＝Q_t-L_tt∈T

in the formula: m represents a system generator set, M represents a generator serial number, B_mRepresenting the state of the generator, normal operation or fault not-operating, G_mRepresenting the generated power of generator m, Q_tRepresenting the load on node t;

-F_l ^max≤F_l≤F_l ^maxl∈W

in the formula, F_l ^maxRepresenting the transmission limit of the line l, G_m ^minAnd G_m ^maxRepresenting the maximum and minimum output of the generator m, respectively.

Further, for the power transmission network, the state space of the power system attack success is firstly traversed partially and then traversed by sampling, and the method comprises the following steps:

multiplying the probabilities of the line states of all permutation and combination in the power system attack success state space respectively to obtain the occurrence probability of all the states in the state space, and sequencing the occurrence probability;

traversing the line states from front to back according to the sequencing result according to the set number of the states needing to be traversed;

and randomly sampling the rest line states according to the set number of the states needing to be sampled.

Further, the sampling method adopts a Monte Carlo method.

Further, the network attack types include a spurious data injection attack against power system measurements, a replay attack against power system instructions, and a threshold modification attack against power system protection devices. .

The invention achieves the following beneficial effects: according to the method, the line attack success probability is calculated according to the attack types and the attack line number, the network attack success state space of the power system is obtained, the state space is traversed, the load loss amount is calculated respectively, the expected fault set of the power system considering the network attack is obtained, the expected fault set in the existing three defense lines is supplemented, the appropriate expected fault can be matched for a dispatcher, and the subsequent control effect is improved.

Drawings

FIG. 1 is a flow chart of a method for generating an expected failure set according to an embodiment of the present invention;

FIG. 2 is a two-machine four-node system diagram in accordance with an embodiment of the present invention;

fig. 3 is a diagram of IEEE57 wiring system in accordance with an embodiment of the present invention.

Detailed Description

The invention is further described below with reference to the accompanying drawings. The following examples are only for illustrating the technical solutions of the present invention more clearly, and the protection scope of the present invention is not limited thereby.