CN112541679A - Protection method for power grid under heavy load distribution attack - Google Patents

Protection method for power grid under heavy load distribution attack Download PDF

Info

Publication number
CN112541679A
CN112541679A CN202011466030.3A CN202011466030A CN112541679A CN 112541679 A CN112541679 A CN 112541679A CN 202011466030 A CN202011466030 A CN 202011466030A CN 112541679 A CN112541679 A CN 112541679A
Authority
CN
China
Prior art keywords
load
power grid
attack
node
threshold value
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202011466030.3A
Other languages
Chinese (zh)
Inventor
梁毅
张明理
宋卓然
孙秋野
韩震焘
刘鑫蕊
李华
李上来
张娜
武志锴
尹婧娇
孙岩
马强
李美君
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
State Grid Corp of China SGCC
Economic and Technological Research Institute of State Grid Liaoning Electric Power Co Ltd
Original Assignee
State Grid Corp of China SGCC
Economic and Technological Research Institute of State Grid Liaoning Electric Power Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by State Grid Corp of China SGCC, Economic and Technological Research Institute of State Grid Liaoning Electric Power Co Ltd filed Critical State Grid Corp of China SGCC
Priority to CN202011466030.3A priority Critical patent/CN112541679A/en
Publication of CN112541679A publication Critical patent/CN112541679A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
    • G06Q10/06312—Adjustment or analysis of established resource schedule, e.g. resource or task levelling, or dynamic rescheduling
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/24—Querying
    • G06F16/245—Query processing
    • G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
    • G06F16/2462—Approximate or statistical queries
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/24—Querying
    • G06F16/245—Query processing
    • G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
    • G06F16/2465—Query processing support for facilitating data mining operations in structured databases
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/24—Querying
    • G06F16/245—Query processing
    • G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
    • G06F16/2477—Temporal data queries
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/06—Energy or water supply
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T11/00—2D [Two Dimensional] image generation
    • G06T11/20—Drawing from basic elements, e.g. lines or circles
    • G06T11/203—Drawing of straight lines or curves

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Human Resources & Organizations (AREA)
  • General Physics & Mathematics (AREA)
  • Economics (AREA)
  • Strategic Management (AREA)
  • Probability & Statistics with Applications (AREA)
  • Databases & Information Systems (AREA)
  • Software Systems (AREA)
  • Tourism & Hospitality (AREA)
  • Computational Linguistics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • General Business, Economics & Management (AREA)
  • General Engineering & Computer Science (AREA)
  • Marketing (AREA)
  • Fuzzy Systems (AREA)
  • Data Mining & Analysis (AREA)
  • Mathematical Physics (AREA)
  • Development Economics (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Health & Medical Sciences (AREA)
  • Game Theory and Decision Science (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Educational Administration (AREA)
  • Public Health (AREA)
  • Water Supply & Treatment (AREA)
  • General Health & Medical Sciences (AREA)
  • Primary Health Care (AREA)
  • Supply And Distribution Of Alternating Current (AREA)

Abstract

The invention provides a protection method of a power grid under the attack of load distribution, and relates to the technical field of power transmission and distribution. The method comprises the steps of firstly, describing a historical load curve of each time interval of a power grid through analysis of historical data of the power grid, and predicting load electricity utilization conditions and load electricity utilization behaviors of each time interval of the future power grid to obtain a power grid load prediction curve; determining the range of the load fluctuation threshold value measured by the intelligent ammeter according to the historical load curve and the load prediction curve of the power grid, and obtaining the predicted load fluctuation threshold value; and then judging whether the threshold value of the load fluctuation measured by the intelligent ammeter at the power grid user side in real time is larger than the predicted load fluctuation threshold value, if the threshold value of the load fluctuation is larger than the predicted load fluctuation threshold value, judging that the intelligent ammeter is attacked by load distribution, calculating the risk probability of the intelligent ammeter being attacked by load distribution, and setting the weight coefficient of the load before and after the attack to guide power grid scheduling personnel to carry out reasonable load distribution.

Description

Protection method for power grid under heavy load distribution attack
Technical Field
The invention relates to the technical field of power transmission and distribution, in particular to a protection method of a power grid under the attack of load distribution.
Background
Network attacks may have catastrophic effects on the grid, and in fact, network attacks directed to the energy sector occur at times. In the future, with the increase of network devices and the reduction of network attack thresholds in the power system, network attacks on the power system will become more frequent and more destructive. Therefore, improving the reliability of the power grid to prevent potential network attacks has become an urgent issue to be addressed and solved in the process of building smart power grids. There are several organizations and groups that research the defense methods and strategies against false data injection attacks.
The load redistribution attack is also a false data injection attack, and a common attack object of the load redistribution is an intelligent electric meter positioned at a user side, and although the intelligent electric meter adopts a strict access control mechanism and a secure communication mechanism, the security vulnerability of an operating system for controlling the intelligent electric meter is still difficult to guarantee. Malignant data invading the intelligent electric meter can be transmitted to other intelligent electric meters, the batch control switch is disconnected to cause the accident that power grid users quit power grid interaction, and the electric energy metering value can be tampered, so that the analysis and decision making errors and direct economic losses of a power grid company are caused.
The dummy data satisfies the following condition, detects and changes the result of the state estimation by the bad data.
ΔZ＝HΔX
Wherein, Δ Z represents a false data vector injected by an attacker in a power grid measurement value, H is a measured jacobian matrix of the power grid, and Δ X represents an error vector introduced in a state variable due to the injection of the false data. In the dc state estimation, the measured values of the grid mainly include the output active power of the generator, the load power, and the branch active power. Not all measurement information can be attacked and tampered, and some key measurement information has strict protection measures and is difficult to attack. There is usually direct real-time communication between the power plant and the control center, which makes it easy to discover the output active power of a tampered generator. Generally, the output active power of the generator is considered to be not attacked, and then a spurious data injection attack model (i.e. heavy load distribution attack) shown as the following formula is obtained:
Figure BDA0002834219630000011
wherein, Δ PFIs the active measurement value of the injected branch, Δ PDIs the injected load measurement, SF is the transfer factor matrix derived from the grid network topology and electrical parameters, BL is the bus load correlation matrix (determined by the load position), I is the unit matrix.
Considering that the injected load measurements should be within a certain range, the above attack model can be further expressed as:
Figure BDA0002834219630000012
-τPD,i≤ΔPD,i≤τPD,i (2)
ΔPF＝-SF×BL×ΔPD (3)
wherein, BDRepresents the loaded bus in the power grid, and tau represents the range limit of the attack.
(1) The equation represents that the sum of the injected attack amount of the load is 0, otherwise the probability that the attack is detected is greatly increased due to the fact that the generator and load data in the measurement data are not conserved. (2) The formula shows that the injection attack amount of the load should be within a certain range, and the formula (3) shows that the corresponding branch power flow changes along with the change of the load. It should be noted that in a heavy load distribution attack, the actual load and branch flow are not changed, but the measurement data received by the control center is tampered by the attacker.
Although a heavy load distribution attack cannot directly destroy the physical state like a load change attack, a line attack, and the like, it can have a serious influence in an indirect manner such as misleading the control center to make an incorrect power schedule.
Therefore, the protection strategy of interaction between the electric automobile and the power grid under the attack of heavy load distribution has certain theoretical basis and practical significance.
Disclosure of Invention
The technical problem to be solved by the invention is to provide a protection method of a power grid under the attack of load redistribution aiming at the defects of the prior art, and to realize the protection of the power grid aiming at the attack of load redistribution.
In order to solve the technical problems, the technical scheme adopted by the invention is as follows: a protection method for a power grid under a heavy load distribution attack comprises the following steps:
step 1: based on a power big data descriptive analysis method and a predictive analysis method, historical load curves of each time interval of a power grid are described through analysis of historical data of the power grid, and load electricity utilization conditions and load electricity utilization behaviors of each time interval of the future power grid are predicted to obtain a power grid load prediction curve;
the following steps are carried out on the load data of all m users in a week from the historical data set of the power grid to extract the electricity utilization behavior characteristics of all the users in the current month:
1) dividing each day into 24 time periods, and calculating the sum of the power consumption of the power grid load in each time period;
2) calculating the average electricity consumption of each user in each day and each time period in a week;
forming a power grid historical load curve based on statistical analysis and big data analysis technologies, and predicting the power grid load based on the historical load curve to form a load prediction curve;
step 2: determining the range of the measured load fluctuation threshold value of the intelligent ammeter according to the historical load curve and the load prediction curve of the power grid, then determining the future load fluctuation threshold value through Monte Carlo simulation, and taking the mean value of simulation results as the most possible value of the load fluctuation threshold value, namely the predicted load fluctuation threshold value;
and step 3: judging whether the intelligent electric meter at the user side of the power grid is attacked by load distribution;
judging whether a load fluctuation threshold value measured by a smart meter at a power grid user side in real time exceeds a predicted load fluctuation threshold value, if so, judging that the smart meter suffers from a heavy load distribution attack, and executing the step 4, and if not, judging that the smart meter does not suffer from the heavy load distribution attack;
and 4, step 4: scheduling the power grid load under the attack of load distribution; calculating the probability of the attack risk of the intelligent electric meter subjected to the load distribution, and setting the weight coefficient of the load before and after the attack to guide power grid dispatching personnel to carry out reasonable load distribution;
step 4.1: setting measurement information of a power grid defender selecting and protecting some intelligent electric meters, and an attacker selecting and attacking some intelligent electric meters; with AaRepresents an attack strategy that may be selected by an attacker,
Figure BDA0002834219630000031
Figure BDA0002834219630000032
n is selected from the whole m intelligent electric meters in the power gridaThe number of the selectable strategies of all possible combinations of the attack targets is
Figure BDA0002834219630000033
Likewise, with AdRepresents a defense strategy that is selectable by a defender,
Figure BDA0002834219630000034
Figure BDA0002834219630000035
n is selected from the whole m intelligent electric meters in the power griddAll possible combinations of the target as defense, the number of the selectable strategies is
Figure BDA0002834219630000036
Step 4.2: the economic loss of the power grid caused by the load redistribution attack is expressed as the increment of the operation cost of the power grid, and the following formula is shown:
Figure BDA0002834219630000037
wherein M is the increment of the running cost of the power grid, cgIn order to account for the current operating costs of the grid,
Figure BDA0002834219630000038
the current power consumption of the power grid is used,
Figure BDA0002834219630000039
is the load switching quantity of the ith node in the power grid, BGIs a collection of grid load nodes, BDCS is the cost of load switching, and CS is the set of nodes corresponding to the attacked smart meternormalThe operation cost of the power grid system during normal operation is represented;
step 4.3: determining an objective function of a defender;
for m attack targets, let DCmTo reduce the minimum attack investment for attacking these target success rates, D is satisfied when defending these target resourcesm≥DCmProbability of success of attack qm(Dm) As shown in the following equation:
Figure BDA00028342196300000310
wherein, for the mth physical target, DmTo allocate defensive resources to these physical targets, αm＝-ln(qm0)/DCm，qm0The attack success rate at the lowest defense investment is achieved;
when D is presentn＜DCmProbability p of success of attackm(Am)＝1，AmA set of m smart meters selected for an aggressor;
in the actual power grid planning, the defense resources are limited, and the total quantity of the power grid defense resources is DtotalThen, there are:
∑Dm≤Dtotal
attack on the gridThe post-synthetic loss is
Figure BDA00028342196300000311
The objective function of the defender to cut the amount of load is then:
min max{qmPC}
wherein q ismFor the attack success probability under this defense strategy, PCSelecting an optimal attack scheme for an attacker under the defense strategy, and adopting the load reduction total amount under the optimal countermeasure after the attacker suffers from the attack;
after the power grid is attacked, load distribution measures are taken to prevent the power grid from generating cascading failures and ensure the stable operation of the power grid; through reasonable load distribution, the loss caused by the attack is minimized, and the objective function of the defensive party on the increment of the power grid operation cost is obtained as follows:
minM
step 4.4: the method comprises the following steps that an attacked node in a power grid distributes loads by adopting a local load distribution strategy nearby;
if the node i fails due to attack, the load of the node i is completely shared by the neighbor nodes thereof according to the following formula, and the proportion of the load received by one neighbor node j is as follows:
Figure BDA0002834219630000041
wherein λ isjRepresenting the proportion of the load received by node j, CjBeing the capacity of node j, ΓiA set of neighbor nodes representing node i;
the load DeltaL distributed to the node j after the node i fails due to attackijComprises the following steps:
Figure BDA0002834219630000042
wherein L isiRepresenting the load borne by the node i when the node i fails;
step 4.5: the load is transferred and distributed based on the method of weighting and matching the original load data and the current load data, and the following formula is shown:
Figure BDA0002834219630000043
wherein S isijThe amount of load actually allocated by node i to neighbor node j for the dispatcher,
Figure BDA0002834219630000044
obtaining the mean value of the original load of the node i based on big data and statistical analysis, wherein alpha and beta are weighting coefficients; and if the fluctuation of the load quantity of the attacked node is larger than the load fluctuation threshold determined in the step 2, reducing the alpha value and increasing the beta value.
Adopt the produced beneficial effect of above-mentioned technical scheme to lie in: the invention provides a method for protecting a power grid under heavy load distribution attack, which further improves the influence of the load redistribution attack on the power grid dispatching process, introduces a big data description method and a prediction method to describe and predict a power grid load curve, provides a method for judging whether the power grid is under the heavy load distribution attack by taking a power grid load fluctuation threshold value as a judgment result, introduces a heavy load distribution attack model, and runs the economic cost of the power grid after the power grid is attacked, in order to reduce the operation cost of the power grid after the heavy load distribution attack to the maximum extent, a strategy of local load redistribution is provided, in view of the deceptiveness of the heavy load distribution attack, a weighting coefficient is introduced, the average value of the historical load of the power grid is included in the reference of the load redistribution of the adjacent nodes, the influence of the false data on the load redistribution of the power grid is effectively weakened, and the stability and the reliability of the power grid are obviously improved.
Drawings
Fig. 1 is a topology diagram of an IEEE14 power system according to an embodiment of the present invention;
fig. 2 is a flowchart of a protection method for a power grid under a heavy load distribution attack according to an embodiment of the present invention;
fig. 3 is a schematic diagram of a nearby local load reallocation strategy according to an embodiment of the present invention.
Detailed Description
The following detailed description of embodiments of the present invention is provided in connection with the accompanying drawings and examples. The following examples are intended to illustrate the invention but are not intended to limit the scope of the invention.
In this embodiment, an IEEE 14-bus system is taken as an example, and the protection method for the power grid under the heavy load distribution attack is adopted to protect the interaction process between the electric vehicle and the power grid.
In the present embodiment, the electric vehicle is a load of the IEEE14 power system shown in fig. 1, the system includes 14 bus bars, and for convenience of analysis, the loads of the bus bars 13 and 14 are increased. The transmission line 1 has a capacity of 160MW and the remaining lines have a capacity of 60 MW. Suppose that each bus is provided with an intelligent electric meter as a measuring unit, and each branch is provided with an intelligent electric meter as a measuring unit. The generator related parameters are shown in table 1, and the bus node parameters are shown in table 2. Simulation analysis was performed using MATLAB R2018 b.
TABLE 1 Generator parameters
Bus where generator is located 1 2 3 6 8
Minimum power (MW) 0 0 0 0 0
Maximum power (MW) 332.4 140 100 100 100
Cost of electricity generation ($/MWh) 20 30 40 50 35
TABLE 2 bus parameters
Bus bar Load (MW) Bus bar Load (MW)
1 0 8 0
2 21.7 9 29.5
3 94.2 10 9
4 47.8 11 3.5
5 7.6 12 6.1
6 11.2 13 53.5
7 0 14 54.9
The normal optimal economic dispatch of the power system in this embodiment is shown in table 3, where the power generation amount of each generator satisfies the power limit, the total power generation amount satisfies the load demand, and the branch power flow also satisfies the capacity limit, and at this time, the system operation cost is the lowest.
TABLE 3 comparison of economic dispatch under the most destructive attack with Normal economic dispatch
Figure BDA0002834219630000061
In this embodiment, a method for protecting a power grid under a heavy load distribution attack, as shown in fig. 2, includes the following steps:
step 1: describing historical load curves of the electric vehicle in all periods of the power grid by analyzing historical data of the power grid based on a power big data descriptive analysis method and a predictive analysis method, and predicting the power utilization condition of the electric vehicle and the power utilization behavior of the electric vehicle in all periods of the power grid in the future to obtain a load prediction curve of the electric vehicle;
the following steps are carried out on all m user load data in a week unit from the historical data set of the power grid to extract the electricity utilization behavior characteristics of all users in the current month:
1) dividing each day into 24 time periods, and calculating the sum of the power consumption of the power grid load in each time period;
2) calculating the average electricity consumption of each user in each day and each time period in a week;
forming an electric vehicle historical load curve based on statistical analysis and big data analysis technologies, and predicting the electric vehicle load based on the electric vehicle historical load curve to form an electric vehicle load prediction curve;
step 2: determining the range of the load fluctuation threshold value measured by the intelligent ammeter according to the historical load curve and the load prediction curve of the power grid, then determining the future load fluctuation threshold value of the electric automobile through Monte Carlo simulation, and taking the mean value of the simulation results as the most possible value of the load fluctuation threshold value, namely the predicted load fluctuation threshold value;
and step 3: judging whether the intelligent electric meter at the user side of the power grid is attacked by load distribution;
judging whether a load fluctuation threshold value measured by a smart meter at a power grid user side in real time exceeds a predicted load fluctuation threshold value, if so, judging that the smart meter suffers from a heavy load distribution attack, and executing the step 4, and if not, judging that the smart meter does not suffer from the heavy load distribution attack;
and 4, step 4: scheduling the power grid load under the attack of load distribution; calculating the probability of the attack risk of the intelligent electric meter subjected to the load distribution, and setting the weight coefficient of the load before and after the attack to guide power grid dispatching personnel to carry out reasonable load distribution;
step 4.1: setting measurement information of a power grid defender selecting and protecting some key intelligent electric meters, and an attacker selecting and attacking some intelligent electric meters; with AaRepresents an attack strategy that may be selected by an attacker,
Figure BDA0002834219630000071
Figure BDA0002834219630000072
n is selected from the whole m intelligent electric meters in the power gridaThe number of the selectable strategies of all possible combinations of the attack targets is
Figure BDA0002834219630000073
Likewise, with AdRepresents a defense strategy that is selectable by a defender,
Figure BDA0002834219630000074
Figure BDA0002834219630000075
n is selected from the whole m intelligent electric meters in the power griddAll possible combinations of the target as defense, the number of the selectable strategies is
Figure BDA0002834219630000076
Step 4.2: the strategy of an attacker is to bring the maximum damage to the power grid and enable the economic loss of the power grid to reach the maximum, and the control strategy of a defender is to reduce the economic loss of the power grid to the minimum; therefore, the economic loss of the power grid caused by the load redistribution attack is expressed as an increase of the operation cost of the power grid, and the following formula is shown:
Figure BDA0002834219630000077
wherein M is the increment of the running cost of the power grid, cgIn order to account for the current operating costs of the grid,
Figure BDA0002834219630000078
the current power consumption of the power grid is used,
Figure BDA0002834219630000079
is the load switching quantity of the ith node in the power grid, BGIs a collection of grid load nodes, BDCS is the cost of load switching, and CS is the set of nodes corresponding to the attacked smart meternormalThe running cost of the power grid system during normal running (not attacked);
without considering the attack limits, the most destructive attacks that the attacker can implement are obtained (maximizing the increase in the operating cost of the power system). Table 4 shows the attack strategy of the measurement unit that the attacker needs to control and tamper with when implementing the attack, i.e. the attacker.
TABLE 4 measurement information to be tampered with in the most destructive heavy load distribution attack
Figure BDA00028342196300000710
Figure BDA0002834219630000081
Step 4.2: determining an objective function of a defender;
for m attack targets, let DCmTo reduce the minimum attack investment for attacking these target success rates, D is satisfied when defending these target resourcesm≥DCmProbability of success of attack qm(Dm) As shown in the following equation:
Figure BDA0002834219630000082
wherein, for the mth physical target, DmTo allocate defensive resources to these physical targets, αm＝-ln(qm0)/DCm，qm0The attack success rate at the lowest defense investment is achieved;
when D is presentm＜DCmProbability p of success of attackm(Am)＝1，AmA set of m smart meters selected for an aggressor;
in the actual power grid planning, the defense resources are limited, and the total quantity of the power grid defense resources is DtotalThen, there are:
ΣDm≤Dtotal
the comprehensive loss of the power grid after being attacked is
Figure BDA0002834219630000083
The objective function of the defender to cut the amount of load is then:
min max{qmPC}
wherein q ismFor the attack success probability under this defense strategy, PCSelecting an optimal attack scheme for an attacker under the defense strategy, and adopting the load reduction total amount under the optimal countermeasure after the attacker suffers from the attack;
after the power grid is attacked, load distribution measures are taken by adjusting the load of the electric automobile so as to prevent the power grid from generating cascading failures and ensure the stable operation of the power grid; through reasonable load distribution, the loss caused by the attack is minimized, and the objective function of the defensive party on the increment of the power grid operation cost is obtained as follows:
minM
considering the optimal defense strategy of attack and defense limitation, in order to defend the heavy load distribution attack, a defender selects some key measuring units for protection. The IEEE 14-bus system comprises 20 branch load flow measuring units and 11 load measuring units. Due to limited resources and budgets, defenders can only protect some of the measurement units.
Step 4.3: an attacked node in a power grid distributes loads by adopting a local load distribution strategy nearby, as shown in FIG. 3;
when one or more nodes in the power system network fail, the load of the failed node is transferred to other nodes, and the load of the failed node is distributed to all neighbor nodes of the node and nodes directly connected with the node according to the capacity proportion of the neighbor nodes; if the node i fails due to attack, the load of the node i is completely shared by the neighbor nodes thereof according to the following formula, and the proportion of the load received by one neighbor node j is as follows:
Figure BDA0002834219630000084
wherein λ isjRepresenting the proportion of the load received by node j, CjBeing the capacity of node j, ΓiA set of neighbor nodes representing node i;
the load DeltaL distributed to the node j after the node i fails due to attackijComprises the following steps:
Figure BDA0002834219630000091
wherein L isiRepresenting the load borne by the node i when the node i fails;
step 4.4: because of the influence of the load redistribution attack, the measurement information of the smart electric meter is not accurate at the moment and cannot reflect the real running state of the power grid, if the dispatcher redistributes the load according to the measurement information received at the moment, the intention of an attacker can be corrected, and the load is transferred and distributed based on a method of weighted matching of original load data and current load data, wherein the following formula is shown:
Figure BDA0002834219630000092
wherein S isijThe amount of load actually allocated by node i to neighbor node j for the dispatcher,
Figure BDA0002834219630000093
obtaining the mean value of the original load of the node i based on big data and statistical analysis, wherein alpha and beta are weighting coefficients; if the fluctuation of the load of the attacked node exceeds the load fluctuation threshold determined in the step 2, the alpha value is reduced, the beta value is increased, namely the weight occupied by the historical load statistical data is increased, the error redistribution of a large number of false loads by a dispatcher is reduced, and the operation cost is reduced.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; such modifications and substitutions do not depart from the spirit of the corresponding technical solutions and scope of the present invention as defined in the appended claims.

Claims (3)

1. A protection method of a power grid under a heavy load distribution attack is characterized in that: the method comprises the following steps:
step 1: based on a power big data descriptive analysis method and a predictive analysis method, historical load curves of each time interval of a power grid are described through analysis of historical data of the power grid, and load electricity utilization conditions and load electricity utilization behaviors of each time interval of the future power grid are predicted to obtain a power grid load prediction curve;
step 2: determining the range of the measured load fluctuation threshold value of the intelligent ammeter according to the historical load curve and the load prediction curve of the power grid, then determining the future load fluctuation threshold value through Monte Carlo simulation, and taking the mean value of simulation results as the most possible value of the load fluctuation threshold value, namely the predicted load fluctuation threshold value;
and step 3: judging whether the intelligent electric meter at the user side of the power grid is attacked by load distribution;
judging whether a load fluctuation threshold value measured by a smart meter at a power grid user side in real time is larger than a predicted load fluctuation threshold value or not, if the predicted load fluctuation threshold value is exceeded, judging that the smart meter is attacked by heavy load distribution, and executing the step 4, if the predicted load fluctuation threshold value is not exceeded, judging that the smart meter is not attacked by heavy load distribution;
and 4, step 4: scheduling the power grid load under the attack of load distribution; and calculating the probability of the attack risk of the intelligent electric meter subjected to the load distribution, and setting the weight coefficient of the load before and after the attack to guide the power grid dispatching personnel to carry out reasonable load distribution.
2. The method for protecting a power grid from a heavy load distribution attack as claimed in claim 1, wherein: the specific method of the step 1 comprises the following steps:
the following steps are carried out on the load data of all m users in a week from the historical data set of the power grid to extract the electricity utilization behavior characteristics of all the users in the current month:
1) dividing each day into 24 time periods, and calculating the sum of the power consumption of the power grid load in each time period;
2) calculating the average electricity consumption of each user in each day and each time period in a week;
and forming a power grid historical load curve based on statistical analysis and big data analysis technologies, and predicting the power grid load based on the historical load curve to form a load prediction curve.
3. The method for protecting a power grid from a heavy load distribution attack as claimed in claim 1, wherein: the specific method of the step 4 comprises the following steps:
step 4.1: setting measurement information of a power grid defender selecting and protecting some intelligent electric meters, and an attacker selecting and attacking some intelligent electric meters; with AaRepresents an attack strategy that may be selected by an attacker,
Figure FDA0002834219620000011
Figure FDA0002834219620000012
n is selected from the whole m intelligent electric meters in the power gridaThe number of the selectable strategies of all possible combinations of the attack targets is
Figure FDA0002834219620000013
Likewise, with AdRepresents a defense strategy that is selectable by a defender,
Figure FDA0002834219620000014
Figure FDA0002834219620000015
n is selected from the whole m intelligent electric meters in the power griddAll possible combinations of the target as defense, the number of the selectable strategies is
Figure FDA0002834219620000016
Step 4.2: the economic loss of the power grid caused by the load redistribution attack is expressed as the increment of the operation cost of the power grid, and the following formula is shown:
Figure FDA0002834219620000021
wherein M is the increment of the running cost of the power grid, cgIn order to account for the current operating costs of the grid,
Figure FDA0002834219620000022
the current power consumption of the power grid is used,
Figure FDA0002834219620000023
is the load switching quantity of the ith node in the power grid, BGIs a collection of grid load nodes, BDCS is the cost of load switching, and CS is the set of nodes corresponding to the attacked smart meternormalThe operation cost of the power grid system during normal operation is represented;
step 4.3: determining an objective function of a defender;
for m attack targets, let DCmTo reduce the minimum attack investment for attacking these target success rates, D is satisfied when defending these target resourcesm≥DCmProbability of success of attack qm(Dm) As shown in the following equation:
Figure FDA0002834219620000024
wherein, for the mth physical target, DmTo allocate defensive resources to these physical targets, αm＝-ln(qm0)/DCm，qm0The attack success rate at the lowest defense investment is achieved;
when D is presentm＜DCmProbability p of success of attackm(Am)＝1，AmA set of m smart meters selected for an aggressor;
in the actual power grid planning, the defense resources are limited, and the total quantity of the power grid defense resources is DtotalThen, there are:
ΣDm≤Dtotal
the comprehensive loss of the power grid after being attacked is
Figure FDA0002834219620000025
The objective function of the defender to cut the amount of load is then:
min max{qmPC}
wherein q ismFor the attack success probability under this defense strategy, PCSelecting an optimal attack scheme for an attacker under the defense strategy, and adopting the load reduction total amount under the optimal countermeasure after the attacker suffers from the attack;
after the power grid is attacked, load distribution measures are taken to prevent the power grid from generating cascading failures and ensure the stable operation of the power grid; through reasonable load distribution, the loss caused by the attack is minimized, and the objective function of the defensive party on the increment of the power grid operation cost is obtained as follows:
min M
step 4.4: the method comprises the following steps that an attacked node in a power grid distributes loads by adopting a local load distribution strategy nearby;
if the node i fails due to attack, the load of the node i is completely shared by the neighbor nodes thereof according to the following formula, and the proportion of the load received by one neighbor node j is as follows:
Figure FDA0002834219620000026
wherein λ isjRepresenting the proportion of the load received by node j, CjBeing the capacity of node j, ΓiA set of neighbor nodes representing node i;
the load DeltaL distributed to the node j after the node i fails due to attackijComprises the following steps:
Figure FDA0002834219620000031
wherein L isiRepresenting the load borne by the node i when the node i fails;
step 4.5: then, load transfer and distribution are carried out based on a method of weighting and matching the original load data and the current load data, and the following formula is shown:
Figure FDA0002834219620000032
wherein S isijThe amount of load actually allocated by node i to neighbor node j for the dispatcher,
Figure FDA0002834219620000033
obtaining the mean value of the original load of the node i based on big data and statistical analysis, wherein alpha and beta are weighting coefficients;
and if the fluctuation of the load quantity of the attacked node is larger than the load fluctuation threshold determined in the step 2, reducing the alpha value and increasing the beta value.
CN202011466030.3A 2020-12-14 2020-12-14 Protection method for power grid under heavy load distribution attack Pending CN112541679A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011466030.3A CN112541679A (en) 2020-12-14 2020-12-14 Protection method for power grid under heavy load distribution attack

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011466030.3A CN112541679A (en) 2020-12-14 2020-12-14 Protection method for power grid under heavy load distribution attack

Publications (1)

Publication Number Publication Date
CN112541679A true CN112541679A (en) 2021-03-23

Family

ID=75018545

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011466030.3A Pending CN112541679A (en) 2020-12-14 2020-12-14 Protection method for power grid under heavy load distribution attack

Country Status (1)

Country Link
CN (1) CN112541679A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113435923A (en) * 2021-06-15 2021-09-24 北京百度网讯科技有限公司 Power consumption prediction method and device and electronic equipment
CN113643151A (en) * 2021-08-02 2021-11-12 广西大学 Information physical cooperation load redistribution attack method and system

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113435923A (en) * 2021-06-15 2021-09-24 北京百度网讯科技有限公司 Power consumption prediction method and device and electronic equipment
CN113643151A (en) * 2021-08-02 2021-11-12 广西大学 Information physical cooperation load redistribution attack method and system
CN113643151B (en) * 2021-08-02 2023-05-09 广西大学 Information physical collaboration load redistribution attack method and system

Similar Documents

Publication Publication Date Title
Taheri et al. Distribution system resilience enhancement via mobile emergency generators
El-Zonkoly Optimal placement of multi-distributed generation units including different load models using particle swarm optimization
Liu et al. Leveraging strategic detection techniques for smart home pricing cyberattacks
CN112688315B (en) Attack and defense system and method based on electric vehicle power distribution network information physical system
Choeum et al. OLTC-induced false data injection attack on volt/VAR optimization in distribution systems
CN112541679A (en) Protection method for power grid under heavy load distribution attack
Jalilpoor et al. Network hardening and optimal placement of microgrids to improve transmission system resilience: A two-stage linear program
Ghasemi et al. A stochastic planning model for improving resilience of distribution system considering master-slave distributed generators and network reconfiguration
Lai et al. Sizing battery storage for islanded microgrid systems to enhance robustness against attacks on energy sources
Liu et al. A robust strategy for leveraging soft open points to mitigate load altering attacks
Le et al. Distributionally robust chance constrained planning model for energy storage plants based on Kullback–Leibler divergence
Guo et al. Reinforcement-learning-based dynamic defense strategy of multistage game against dynamic load altering attack
Lai et al. A resilient power system operation strategy considering transmission line attacks
Liu et al. Reinforcement learning for cyber-physical security assessment of power systems
Faried et al. Probabilistic technique for sizing FACTS devices for steady-state voltage profile enhancement
Liu et al. A distributionally robust scheme for critical component identification to bolster cyber-physical resilience of power systems
Wang et al. A clustering-based framework for searching vulnerabilities in the operation dynamics of Cyber-Physical Energy Systems
Du et al. Hybrid robust tri-level defense model against multiperiod uncertain attacks
CN113065218B (en) Electric power system reliability evaluation method, device and system considering LR attack
Khan et al. Cyber security of market-based congestion management methods in power distribution systems
Zhou et al. Active network management considering wind and load forecasting error
Koutsoukis et al. Active distribution network planning based on a hybrid genetic algorithm-nonlinear programming method
Su et al. Optimal defense strategy against load redistribution attacks under attacker’s resource uncertainty: A trilevel optimization approach
Che et al. The impact of ramp-induced data attacks on power system operational security
Fan et al. Research on optimal load shedding for active distribution network based on genetic algorithm

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