CN114512983A - Distributed power supply elastic control method for network attack - Google Patents

Abstract

The invention discloses a distributed power elastic control method for network attack, which belongs to the technical field of microgrid operation control. The method firstly proposes a control protocol to adjust the quality of the communication link of the microgrid; secondly, using the information obtained by each distributed power source from the neighbor nodes, the weighted average subsequence reduction algorithm is used to effectively discard the attacked neighbors. Finally, the distributed secondary control is performed to realize the frequency and voltage recovery of the system, so as to achieve the purpose of isolating the attacked distributed power supply, so as to improve the effect of the safe operation and elasticity of the microgrid. The control method takes into account both normal communication scenarios and network attack scenarios, and has strong robustness, which is of great significance to ensure the safe and stable operation of the microgrid.

Description

A Distributed Power Elasticity Control Method for Network Attacks

technical field

The invention belongs to the technical field of microgrid operation control, and in particular relates to a distributed power supply elastic control strategy and method with robustness to network attacks.

technical background

With the increasing depletion of non-renewable resources such as oil and coal on the earth and the aggravation of environmental pollution, the research and application of integrated energy systems have received extensive attention from scholars and governments around the world, and more and more distributed power sources are connected to the power grid. . A microgrid is a controllable power system capable of supplying its local loads from available distributed power sources. Among them, distributed power sources can be motor types, such as synchronous generators, or inverter types, to facilitate the integration of emerging resources, such as fuel cells, battery energy storage systems, and solar energy.

Microgrids make great use of information and communication technologies, but this in turn exposes them to cyber threats. Microgrid cybersecurity is critical. In a microgrid control system, both control entities and communication entities can be potential targets of cyber threats. False data injection (FDI) attacks target sensors and control decision-making units, thereby corrupting data transmitted over communication links and affecting microgrid data integrity. Denial of service (DoS) attacks threaten the availability of communications system services. FDI attacks can jeopardize the stability of microgrid voltage and frequency, leading to cascading failures and outages for microgrid users, slowing down the response of the DG control system, and synchronizing the DG with values other than the actual voltage and frequency reference values , overloading or violating microgrid equipment thermal limits. Therefore, it is very important to use distributed power elastic control strategies and methods that are robust to network attacks, and play a vital role in ensuring the stability and economic operation of microgrids.

SUMMARY OF THE INVENTION

Aiming at the deficiencies of the prior art, the present invention proposes a distributed power elastic control strategy and method that is robust to network attacks, thereby improving the security of the power system and ensuring the stable operation of the power grid.

The present invention can be achieved through the following technical solutions:

A distributed power elastic control strategy and method with robustness to network attacks, characterized in that the control method comprises the following steps:

Step 1: Obtain the microgrid system model and parameters, including obtaining the communication topology diagram G of the microgrid system and its corresponding Laplace matrix L, the number of distributed power sources N, and the number of attacked distributed power sources n _NC Wait.

Step 2: Perform a droop control on the microgrid, which is realized by the following formula:

In the formula, Pi and Qi represent the output active power and reactive power of the _i -th distributed power _source , respectively, ω _ni and V _ni represent the rated value of the frequency and amplitude of the output AC voltage, respectively, m _Pi and n _Qi respectively Indicates the droop coefficients of P _i and Q _i , ω _i and v _o,magi are the angular frequency and amplitude of the output voltage of the distributed power supply.

The droop factor is calculated proportionally according to the active and reactive power ratings of the distributed power supply, and is determined according to the following formula:

In the formula, P _maxi , Q _maxi and P _maxj , Q _maxj represent the active power and reactive power ratings of the i-th and j-th distributed power sources, respectively.

Step 3: Perform distributed secondary control on the microgrid.

Further, the distributed secondary control of the microgrid includes the following processes:

S1: Each distributed power source exchanges information with its neighbor nodes;

S2: Update the auxiliary angular frequency and voltage amplitude control variables of the distributed power source according to the distributed control protocol;

S3: Carry out the secondary control of the frequency and voltage amplitude of the distributed power supply, and realize the frequency and voltage recovery of the distributed power supply under normal conditions and network attacks.

Further, in the S1, each distributed power source exchanges information with its neighbor nodes, which means that each distributed power source i sends its own angular frequency ω _i and voltage amplitude v _{o, magi} to the distributed power source of the neighbor node. j, simultaneously obtain a series of angular frequencies ω _j and voltage amplitudes v _o,magj from each neighbor node, and sort them according to size.

Further, updating the auxiliary angular frequency and voltage amplitude control variables of the distributed power supply according to the distributed control protocol in S2 includes the following processes:

Set the quality weight of the communication link between distributed power sources, that is, the weight a _ij , as follows:

In the formula, δ _i and δ _j are the power angles of distributed power sources i and j, respectively. R ₁ and R ₂ describe the relative power angle threshold as an indicator reflecting the health of the microgrid control system. When the power angles of the distributed power sources are relatively close, the microgrid is in a healthy operating state in terms of frequency stability. Therefore, this threshold is set to a relatively small value. As the power angle difference increases, the communication link quality is degraded exponentially until the power angle difference is greater than R ₂ , the communication link quality becomes zero, and the information flow between the two distributed power sources is interrupted. γ is a design parameter that adjusts the smoothness and shape of the exponential function, and a _max represents the maximum value of the parameter communication link quality weight;

Compare the sizes of λ ₂ and η × 4n _NC , where λ ₂ is the second largest eigenvalue of the microgrid communication topology graph, and η is a parameter factor less than 1, which can provide sufficient margin for algebraic connectivity to make it Stay above cybersecurity thresholds.

If λ ₂ <η×4n _NC , update the auxiliary angular frequency v _ωi and auxiliary voltage amplitude v _vi control variables of the distributed power supply according to the following distributed control protocol:

_vvi = 0

是一个控制参数，δ_i是分布式电源i的功率角，v₂是对应于特征值λ₂的特征向量。In the formula,

is a control parameter, _δi is the power angle of the distributed power source _i , and v2 is the eigenvector corresponding to the eigenvalue _λ2 .

If λ ₂ ≥η×4n _NC , update the auxiliary angular frequency v _ωi and auxiliary voltage amplitude v _vi control variables of the distributed power generation according to the following distributed control protocol:

In the formula, c _ω and c _v are the frequency control gain and the voltage control gain respectively, and there is only one fixed gain of the distributed power source that is not zero, and g _i ≥ 0. R _ωi describes the updated neighbor set in the distributed frequency control protocol of the ith distributed power source. The adjacency set is generated by comparing ω _j and v _o,magj of adjacent distributed power sources with its own ω _i and v _o,magi , if there are n _NC or more values greater than ω _j , then n _NC values greater than ω _j are discarded. If there are less than n _NC values greater than ω _j , these values are discarded. For values smaller than ω _j , the same procedure is applied to discard the neighboring values of ω _j . _Rvi describes the updated neighbor set in the distributed voltage control protocol of the i-th distributed power source, and is generated in a similar way to R _ωi .

Further, the frequency and voltage amplitude secondary control of the distributed power supply is performed in the S3, which is realized by the following formula:

ω _ni =∫(v _wi )dt

V _ni =∫(v _vi )dt

Through the above process, the distributed secondary control can restore the operating frequency ω _i and the terminal voltage amplitude vo _,magi of the distributed power source to the reference frequency ω _ref and the reference voltage v _ref .

The invention is an anti-attack distributed collaborative control algorithm based on a hidden layer, which can solve the secondary control problem of a microgrid under network attacks. Compared with existing attack-resistant distributed control methods, the proposed controller is more robust and can mitigate the adverse effects of time-dependent FDI attacks on the actuators, sensors, and communication links of the control system, and Robust to state-dependent FDI attacks. Furthermore, the algorithm works even if all distributed power sources and communications are compromised.

Description of drawings

Fig. 1 is the flow chart of the present invention;

Fig. 2 is the island microgrid structure diagram of IEEE 34 bus model;

Figure 3 shows the specifications and parameters of the distributed power supply;

Figure 4 is the load specification parameters;

Fig. 5 is a distributed power communication network diagram;

Fig. 6 is the time chart of the frequency of the distributed power supply and the active power ratio under the control of the present invention;

Fig. 7 is the time chart of the voltage amplitude and reactive power ratio of the distributed power supply under the control of the present invention;

FIG. 8 is a timing chart of the power angle of the distributed power source under the control of the present invention.

Detailed ways

Hereinafter, the present invention will be described in detail with reference to the accompanying drawings to make the objects and effects of the present invention more apparent.

The process of applying a distributed power elastic control strategy robust to network attacks in a microgrid is shown in Figure 1, which includes the following steps:

Step 1): Obtain the microgrid system model and parameters, including obtaining the communication topology diagram G of the microgrid system and its corresponding Laplacian matrix L, the number of distributed power sources N, and the number of attacked distributed power sources n _NC et al.

Step 2) Perform primary droop control on the microgrid according to formula (1) to maintain the power balance of the microgrid:

In the formula, Pi and Qi represent the output active power and reactive power of the _i -th distributed power _source , respectively, ω _ni and V _ni represent the rated value of the frequency and amplitude of the output AC voltage, respectively, m _Pi and n _Qi respectively represents the _Pi and Qi _droop coefficients. ω _i and v _o,magi are the angular frequency and amplitude of the output voltage of the distributed power supply.

The droop coefficient is calculated proportionally according to the active and reactive power ratings of the distributed power supply, and is determined according to formula (2):

In the formula, P _maxi , Q _maxi and P _maxj , Q _maxj represent the active power and reactive power ratings of the i-th and j-th distributed power sources, respectively.

Step 3): perform distributed secondary control on the microgrid, which specifically includes the following processes:

3-1) Each distributed power source exchanges information with its neighbor nodes. The specific implementation method is that the distributed power source i sends its own angular frequency ω _i and voltage amplitude v _{o, magi} to the distributed power source j of the neighbor node, and at the same time from the distributed power source j of the neighbor node. The angular frequency ω _j and the voltage amplitude v o,magj of the neighbor node distributed power source j are obtained, and the angular frequency ω _j and the voltage amplitude v _{o ,magj} of a series of neighbor nodes are sorted according to the size.

3-2) According to formula (3), set the quality weight of the communication link between distributed power sources, that is, the weight a _ij :

In the formula, δ _i and δ _j are the power angles of distributed power sources i and j, respectively. R ₁ and R ₂ describe the relative power angle threshold as an indicator reflecting the health of the microgrid control system. When the power angles of the distributed power sources are relatively close, the microgrid is in a healthy operating state in terms of frequency stability. Therefore, this threshold is set to a relatively small value. As the power angle difference increases, the communication link quality is degraded exponentially until the power angle difference is greater than R ₂ , the communication link quality becomes zero, and the information flow between the two distributed power sources is interrupted. γ is a design parameter that adjusts the smoothness and shape of this exponential function.

3-3) Compare the sizes of λ ₂ and η × 4n _NC , and execute different distributed control protocols according to the results.

where λ2 is the _second largest eigenvalue of the microgrid communication topology graph, and η is a parameter factor less than 1 that can provide enough margin for algebraic connectivity to keep it above the network security threshold.

Case 1: If λ ₂ <η×4n _NC , execute the distributed control protocol according to the following equations (4) and (5) to update the auxiliary angular frequency v _ωi and the auxiliary voltage amplitude v _vi of the distributed power supply:

_vvi = 0 (5)

是一个控制参数，δ_i是分布式电源i的功率角，v₂是对应于特征值λ₂的特征向量。In formula (4),

is a control parameter, _δi is the power angle of the distributed power source _i , and v2 is the eigenvector corresponding to the eigenvalue _λ2 .

Case 2: If λ ₂ ≥η×4n _NC , execute the distributed control protocol according to the following equations (7) (8) to update the auxiliary angular frequency v _ωi and auxiliary voltage amplitude v _vi control variables of the distributed power generation:

In the formula, c _ω and c _v are the frequency control gain and the voltage control gain respectively, and there is only one fixed gain of the distributed power source that is not zero, and g _i ≥ 0. R _ωi describes the updated neighbor set in the distributed frequency control protocol of the ith distributed power source. The adjacency set is generated by comparing ω _j and v _o,magj of adjacent distributed power sources with its own ω _i and v _o,magi , if there are n _NC or more values greater than ω _j , then n _NC values greater than ω _j are discarded. If there are less than n _NC values greater than ω _j , these values are discarded. For values smaller than ω _j , the same procedure is applied to discard the neighboring values of ω _j . _Rvi describes the updated neighbor set in the distributed voltage control protocol of the i-th distributed power source, and is generated in a similar way to R _ωi .

3-4) Secondary control of the frequency and voltage amplitude of the distributed power supply is realized by the following formulas (9) and (10)

ω _ni =∫(v _wi )dt (9)

V _ni =∫(v _vi )dt (10)

The effectiveness of the present invention is proved by simulation experiments.

The simulation is tested using the IEEE 34 bus model. The structure of the island microgrid is shown in Figure 2, which integrates 6 distributed power sources, which are marked as DER in the figure. Figures 3 and 4 provide the specifications for the distributed source and load, respectively. The microgrid operates at a frequency of 60Hz with a nominal line-to-line voltage of 24.9kV. The distributed power source is integrated into the feeder through a Y-Y transformer, with a rated voltage of 480V/24.9kV and a rated power of 400kVA.

The communication network diagram of distributed power supply is shown in Figure 5. Set g ₁ =1, ω _ref =2π×60rad/s. The use of _vref is calculated according to Equation (11).

v _ref =k _p (v _nom -v _c,mag )+k _i ∫(v _nom -v _c,mag )dt (11)

In the formula, the parameters k _p and k _i are set to 0.01 and 10, respectively, v _nom is set to 1 pu, and the control gains c _ω and _cv are set to 40. The parameters in the previous formula (3) are as follows: R1 is set to π/50, R2 is set to π/2, γ is set to 5, and a _max is set to 4.

At the start of the simulation, the microgrid operates under primary droop control. At t=0.6s, the traditional secondary frequency and voltage control works, and the distributed secondary control of the present invention works at t=0.65s. At t=0.6s, an FDI attack is launched on the distributed power source DER6, which affects the voltage and frequency recovery of the microgrid. The frequency of the microgrid and the active power of the distributed power source are shown in Figure 6. The critical bus voltage amplitude and reactive power ratio are shown in Figure 7, respectively. After applying the distributed secondary control of the present invention, the frequency of the distributed power source is restored to 60 Hz, and the active power ratio is synchronized to a common value. In addition, the active power of a DG is distributed according to its active power rating. As shown in FIG. 7 , after applying the distributed secondary control of the present invention, the critical bus voltage amplitude is restored to 1pu, and the reactive power ratio of the distributed power source converges to the value before the FDI attack. Figure 8 shows the power angle variation of the distributed power source, as shown in the figure, after the traditional distributed control takes action, the power angles start to drift from each other due to the presence of false information shared with neighbors, when the distributed two of the present invention is adopted. After the first control, the power angle returns to normal.

It can be seen from the simulation example of this implementation that after the control method of the present invention is adopted, the damage information distributed by the attacked distributed power source is effectively discarded, the frequency and voltage recovery of the system is realized, and both the normal communication scene and the network attack scene are taken into consideration. , thereby improving the operational flexibility of the microgrid. The method proposed by the present invention has good control effect.

Claims (2)

1. A distributed power elastic control method for network attack, characterized in that, the control method comprises the following steps: Step 1: Obtain the microgrid system model and parameters, including obtaining the communication topology diagram G of the microgrid system and its corresponding Laplace matrix L, the number of distributed power sources N, and the number of attacked distributed power sources n _NC ; Step 2: Perform a droop control on the microgrid, which is realized by the following formula:

In the formula, Pi and Qi represent the output active power and reactive power of the _i -th distributed power _source , respectively, ω _ni and V _ni represent the rated value of the frequency and amplitude of the output AC voltage, respectively, m _Pi and n _Qi respectively represents the droop coefficients of P _i and Q _i , ω _i and v _{o, magi} are the angular frequency and amplitude of the output voltage of the distributed power supply; The droop factor is calculated proportionally according to the active and reactive power ratings of the distributed power supply, and is determined according to the following formula:

In the formula, P _maxi , Q _maxi and P _maxj , Q _maxj represent the active power and reactive power ratings of the i-th and j-th distributed power sources, respectively; Step 3: Perform distributed secondary control on the microgrid; Further, the distributed secondary control of the microgrid includes the following processes: S1: Each distributed power source exchanges information with its neighbor nodes; S2: Update the auxiliary angular frequency and voltage amplitude control variables of the distributed power source according to the distributed control protocol; S3: Carry out the secondary control of the frequency and voltage amplitude of the distributed power supply, and realize the frequency and voltage recovery of the distributed power supply under normal conditions and network attacks; The step of updating the auxiliary angular frequency and voltage amplitude control variables of the distributed power supply according to the distributed control protocol in S2 includes the following processes: Set the quality weight of the communication link between distributed power sources, that is, the weight a _ij , as follows:

In the formula, δ _i and δ _j are the power angles of the distributed power sources i and j, respectively; R ₁ and R ₂ describe the relative power angle thresholds, which are used as indicators to reflect the health status of the microgrid control system; as the power angle difference increases, Increase, reduce the quality of the communication link exponentially, until the power angle difference is greater than R ₂ , the quality of the communication link becomes zero, and the information flow between the two distributed power sources is interrupted; γ is a design parameter, adjust the The smoothness and shape of the exponential function, a _max represents the maximum value of the quality weight of the parameter communication link; Compare the sizes of λ2 and η×4n _NCs , where _λ2 is the _second largest eigenvalue of the microgrid communication topology graph, and η is a parameter factor less than 1 that provides enough margin for the algebraic connectivity to keep it above the cybersecurity threshold; If λ ₂ <η×4n _NC , update the auxiliary angular frequency v _ωi and auxiliary voltage amplitude v _vi control variables of the distributed power supply according to the following distributed control protocol:

_vvi = 0 In the formula,

is a control parameter, δ _i is the power angle of the distributed power source i, and v ₂ is the eigenvector corresponding to the eigenvalue λ ₂ ; If λ ₂ ≥η×4n _NC , update the auxiliary angular frequency v _ωi and auxiliary voltage amplitude v _vi control variables of the distributed power generation according to the following distributed control protocol:

In the formula, c _ω and c _v are the frequency control gain and the voltage control gain respectively, and there is only one fixed gain of the distributed power source is not zero, and g _i ≥ 0; R _ωi describes the i-th distributed power source The _{updated adjacency} set _in the distributed frequency control _protocol of the _NC or more values greater than ω _j , then n _NC values greater than ω _j are discarded; if there are less than n _NC values greater than ω _j , these values are discarded; for values less than ω _j , apply The same process is used to discard the adjacent values of ω _j ; R _vi describes the updated neighbor set in the distributed voltage control protocol of the i-th distributed power source, and the generation method is similar to that of R _ωi ; Further, the frequency and voltage amplitude secondary control of the distributed power supply is performed in the S3, which is realized by the following formula: ω _ni =∫(v _wi )dt V _ni =∫(v _vi )dt Through the above process, the distributed secondary control can restore the operating frequency ω _i and the terminal voltage amplitude vo _,magi of the distributed power source to the reference frequency ω _ref and the reference voltage v _ref . 2 . The method for elastic control of distributed power supply for network attacks according to claim 1 , wherein: each distributed power supply exchanges information with neighbor nodes, specifically: each distributed power supply i Send its own angular frequency ω _i and voltage amplitude v _o,magi to the distributed power supply j of neighbor nodes, and obtain a series of angular frequency ω _j and voltage amplitude v _o,magj from each neighbor node at the same time, and carry out according to the size. sort.

Images