Regional distributed energy storage self-healing power grid control method for power system

By establishing an information-physical network and event triggering mechanism in the power system, and using distributed energy storage devices to achieve frequency/voltage self-healing control, the problem of scattered power system scheduling difficulties caused by the scattered distribution of new energy is solved, and the system's self-healing ability is improved and frequency/voltage recovery efficiency.

CN119010084APending Publication Date: 2024-11-22CHINA POWER ENG CONSULTING GRP NEW ENERGY CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411114437.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-14
Publication Date
2024-11-22

AI Technical Summary

Technical Problem

The scattered distribution of the new energy industry leads to difficulties in scheduling the power system, especially when the load changes, the system inertia and damping are insufficient, resulting in voltage and frequency changes, affecting system stability and reliability.

Method used

A method for controlling self-healing grids for regional distributed energy storage in power system is proposed. By establishing an information-physical network, a self-healing grid control strategy based on sag control is designed, combined with event triggering mechanism, and using distributed energy storage equipment to achieve frequency/voltage Rapid recovery, optimize communication strategies between distributed energy storage nodes, and reduce communication costs.

Benefits of technology

It effectively improves the system's self-healing control ability, quickly restores the output frequency and voltage of the distributed generator to the rated value, optimizes the system's frequency/voltage adjustment, reduces communication bandwidth and time, and improves the system's dynamic self-healing characteristics.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119010084A_ABST
    Figure CN119010084A_ABST
Patent Text Reader

Abstract

The invention discloses a regional distributed energy storage self-healing power grid control method for a power system. The method comprises the following steps: establishing an information-physical network corresponding to a regional power grid system; designing a self-healing power grid control strategy based on droop control, and introducing a predictive control strategy based on an event triggering mechanism in combination with an information-physical network; according to the event triggering prediction control strategy, information interaction is carried out on an information network in a distributed energy storage node regional power grid, a virtual inertia correction signal is calculated through a state prediction consistency algorithm, frequency / voltage drop generated by droop control is made up in a distributed mode, and distribution of active power and reactive power among nodes according to requirements is achieved; designing an event trigger state prediction control algorithm; optimizing a communication strategy between the distributed energy storage nodes; the communication strategy of the distributed droop control algorithm is optimized, and the output frequency and voltage of the distributed generator are recovered to rated values under the condition that the communication cost is reduced. The method can effectively improve the self-healing control capability of the system.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of power grid control, and in particular relates to a method for controlling a regional distributed energy storage self-healing power grid in a power system. Background Art

[0002] Traditional power systems are mainly based on synchronous generators, which have large inertia and strong damping capabilities. Therefore, the main state variables in traditional power grids are not affected by general conventional disturbances, or the oscillation process caused by them can be effectively and timely suppressed. These traditional power systems have strong robustness and stability. In the face of the global energy crisis and irreversible climate change, new energy power generation technologies such as photovoltaics and wind power have developed rapidly in the past decade. However, with the access of large-scale new energy power generation equipment to the power grid, the inertia and damping of the system have been greatly reduced, which will lead to large voltage and frequency changes under load changes. The relay protection strategy is prone to malfunction, reducing the reliability and stability of system operation. In recent years, the country has guided the new energy industry to provide a part of the energy storage capacity as the system's primary frequency regulation reserve capacity while ensuring the economic benefits of its full power generation through new energy supporting energy storage equipment and other means. However, the new energy industry is scattered and mostly at the end of the power grid. The supporting energy storage equipment capacity, which usually accounts for 10% to 20% of the capacity of new energy sites, is several megawatts to tens of megawatts. Its scattered and fragmented characteristics are particularly prominent, which brings great difficulties to the dispatch of the power system. As a result, the idea of ​​using new energy and energy storage to provide inertia and damping support for the power grid cannot be realized.

[0003] In the typical structure of a regional distributed power system, a number of loads, synchronous generators, energy storage units or other new energy devices are distributed in the power grid and connected to the power system through a grid-based control method. When the system load fluctuates, the synchronous generator can usually adjust the power output according to its fluctuation to maintain the stable operation of the system. When the proportion of new energy power generation equipment is high and the system inertial damping is weak, the regional distributed power system can use grid-based cluster control to make full use of the distributed energy storage equipment in the system to ensure the stable operation of the system. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide a method for controlling a regional distributed energy storage self-healing power grid in an electric power system, so as to solve the problems in the prior art such as the difficulty in dispatching the electric power system when the new energy industry is scattered. Aiming at the frequency regulation demand of the power grid, a coordinated control strategy of utilizing distributed small-capacity energy storage equipment and combining droop control to participate in the primary frequency regulation of the system is proposed, and a frequency / voltage self-healing control strategy is proposed. Based on a multi-node system and distributed information interaction, the output frequency and voltage of the distributed generator are quickly restored to the rated value while reducing the communication cost through an event trigger mechanism.

[0005] According to the technical solution of the present invention, the present invention provides a method for controlling a regional distributed energy storage self-healing power grid in a power system, comprising the following steps:

[0006] Step S1: Establishing an information-physical network corresponding to the regional power grid system, including a physical network and an information network, wherein the physical network provides power transmission and the information network provides information interaction; the information network includes a regional power grid intra-communication network composed of energy storage nodes of the regional power grid system;

[0007] Step S2: Design a self-healing power grid control strategy based on droop control, combine the information-physical network, and introduce a predictive control strategy based on an event-triggered mechanism; the event-triggered predictive control strategy exchanges information on the information network within the regional power grid of the distributed energy storage node, calculates the virtual inertia correction signal through the state prediction consistency algorithm, and compensates for the frequency / voltage drop caused by the droop control in a distributed manner, and realizes the distribution of active power and reactive power among the distributed energy storage nodes according to demand;

[0008] Step S3: Designing an event-triggered state predictive control algorithm;

[0009] Step S4: Optimizing the communication strategy between distributed energy storage nodes through an event-triggered state predictive control algorithm;

[0010] Step S5: Optimize the communication strategy of the distributed droop control algorithm to restore the output frequency and voltage of the distributed generator to the rated value while reducing the communication cost, that is, make the control error approach zero.

[0011] Furthermore, the physical network includes m network nodes in the regional power grid (RG), and the nodes include thermal power plants, new energy power plants or substations, etc., and the set N RG ={1,2,…,m}, the kth physical network node is represented by RG k , k∈N RG ; Among them, the energy storage devices are distributed in some physical nodes, and the definition includes C distributed energy storage devices DE, and the energy storage device set N DE ={1,2,…,C}, and RG k The i-th distributed energy storage device in is represented as DE i , i∈N DE ; Distributed energy storage equipment DE i The output is based on the measured data of the current node and the feedback data of the surrounding connected nodes.

[0012] Furthermore, in the cyber-physical network, the frequency and voltage of physical nodes are dynamically correlated, and different nodes play different functional roles in the physical network. By abstracting the physical network into a complex network with m nodes and K transmission lines, the nodes include power plants, substations, energy storage, loads, etc. (including at least one of them), and the transmission line branches are the edges in the grid topology. At the same time, the information network constitutes another dimension on this basis. The information network of the physical network nodes in the regional grid (RG) consists of m directed graphs G1, G2, ..., G m Indicates that, respectively, with RG1, RG2, ..., RG m Correspondingly, for the physical node RG k , the communication within the regional power grid is represented by a directed graph G k =(V k ,E k ,A k ) construction, where the non-empty node set V k = {V k1 ,V k2 ,…,V km} represents a set of physical network nodes, node V ki With RG i Corresponding to the edge set Represents the communication line between physical network nodes RG; A k =[a ij ] m×m ,i,j∈N RG is the adjacency matrix, A k The diagonal elements a ii = 0, if and only if there is a communication line between RGi and RGj, the off-diagonal element a ij >0,i≠j, that is, (V i ,V j )∈E k ; Node V i Indegree is defined as Definition G k The in-degree matrix D k and the Laplace matrix L are D k =diag{d in (V i )} and L=D k -A k .

[0013] Furthermore, in the droop control loop, power scheduling is achieved by changing the power-frequency droop coefficient. When the power command issued is P el / Q elWhen the power is dispatched, the power is dispatched and the frequency deviation is reduced by changing the droop power curve; the droop coefficient after adjustment is (the specific method of the self-healing control strategy of droop control is):

[0014]

[0015] Among them, P0 is the rated active power, Q0 is the rated reactive power, P el Node output active power, Q el Output reactive power for the node, is the state prediction control quantity of the node based on event triggering, corresponding to the frequency and voltage values ​​in the power system, L is the system Laplace matrix, k pf is the active power droop coefficient, k qv is the reactive power droop coefficient, and C is the number of regional power grid nodes.

[0016] Furthermore, for the i-th distributed energy storage device DE i , i∈N DE , N DE As a collection of energy storage devices, represents the kth triggering moment of the ith node, Indicates the kth trigger interval; r k represents the event triggering moment of the distributed energy storage system, and L represents the system Laplace matrix; therefore, when the inequality of the following formula 3 is not satisfied, the event triggering executes the communication task and sends a reference signal to other distributed energy storage devices; the event triggering conditions are as follows:

[0017]

[0018] Where ρ and γ are constant parameters, ρ∈(0,1), γ>0, e i (t) represents the state of the i-th agent at the triggering moment With the current state ω i The difference of (t), λ max (L), λ min (L) represents the maximum eigenvalue and minimum eigenvalue of the matrix L, respectively, and z is the trigger condition parameter;

[0019] On this basis, the following state predictor is introduced:

[0020]

[0021] The self-healing control protocol with state predictor is as follows:

[0022]

[0023] Compared with the prior art, the beneficial technical effects of the present invention are as follows:

[0024] 1. The method of the present invention is supported by the information network in the regional power grid composed of RG. By updating the droop control parameters, multiple control objectives are achieved in a distributed manner: frequency / voltage primary regulation and optimization operation of each physical network node. The method has a good optimization and coordination effect and can effectively improve the system's self-healing control capability.

[0025] 2. The present invention proposes a new event-triggered self-healing control algorithm for primary frequency / voltage regulation and optimal output calculation of distributed energy storage power sources. The finite event triggered state controller algorithm can save the bandwidth and time required for communication, accelerate the convergence rate, and meet the convergence performance requirements of the regional power grid system.

[0026] 3. The present invention applies the local control principle. Based on local measured data, the distributed energy storage device uses a local communication strategy to independently resolve the impact of fluctuations by each physical network node, and dynamically adjusts the output of the controllable distributed energy storage power supply. Each physical network node with a distributed energy storage power supply adjusts its output to near the optimal operating point, thereby achieving mutual support of each physical network node against the impact of uncertainty and ensuring that the regional power grid system achieves dynamic self-healing characteristics. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 This is a method control block diagram of an embodiment of the present invention.

[0028] Figure 2 This is a schematic diagram of the information-physical network of the present invention.

[0029] Figure 3 Schematic diagram of a system according to an embodiment of the present invention.

[0030] Figure 4 This is a schematic diagram of the DE node export frequency of the experiment 1 system of an embodiment of the present invention.

[0031] Figure 5 Schematic diagram of DE node output voltage of Experiment 1 system according to an embodiment of the present invention.

[0032] Figure 6 This is a schematic diagram of the DE node export frequency of the Experiment 2 system of an embodiment of the present invention.

[0033] Figure 7 Schematic diagram of DE node output voltage of Experiment 2 system according to an embodiment of the present invention.

[0034] Figure 8 This is a schematic diagram of the DE node export frequency of the experimental third system of an embodiment of the present invention.

[0035] Fig. 9 Schematic diagram of DE node output voltage of Experiment 3 system of an embodiment of the present invention.

[0036] Fig.10 This is a communication diagram of the DE node controller of the Experiment 3 system of an embodiment of the present invention. DETAILED DESCRIPTION

[0037] The present invention provides a method for controlling a regional distributed energy storage self-healing power grid in an electric power system, which is specifically a self-healing power grid control strategy based on droop control and an event trigger mechanism, and solves the problems of difficulty in dispatching the electric power system when the new energy industry is scattered in the prior art. According to the frequency regulation demand of the power grid, a coordinated control strategy of utilizing distributed small-capacity energy storage equipment and combining droop control to participate in the primary frequency regulation of the system is proposed, and a frequency / voltage self-healing control strategy is proposed. Based on a multi-node system and distributed information interaction, the output frequency and voltage of the distributed generator are quickly restored to the rated value while reducing the communication cost through an event trigger mechanism.

[0038] A method for controlling a regional distributed energy storage self-healing power grid in a power system of the present invention comprises the following steps:

[0039] Step S1: Establishing a cyber-physical network corresponding to the regional power grid system, that is, a network integrating information and physics, including a physical network and an information network, wherein the physical network provides power transmission and the information network provides information interaction.

[0040] The information network includes a regional power grid communication network composed of energy storage nodes in the regional power grid system.

[0041] like Figure 2 As shown in the figure, the physical network contains m physical network nodes in the regional power grid (RG), and the m physical network nodes include thermal power plants, new energy power plants or substations, energy storage, loads, etc., and the set N RG ={1,2,…,m}, the kth physical network node is represented by RG k , k∈N RG ; Among them, the distributed energy storage devices are dispersedly arranged in some physical network nodes (ie, in some m nodes), and the definition includes C distributed energy storage devices DE (nodes), and the energy storage device set N DE ={1,2,…,C}, and RG k The i-th distributed energy storage device in is represented as DE i , i∈N DE ; Distributed energy storage equipment DE iThe output is based on the measured data of the current node and the feedback data of the surrounding connected nodes. Furthermore, by abstracting the physical network into a complex network (information-physical network) with m nodes and K transmission lines, the physical network nodes include thermal power plants, new energy power plants or substations, etc., and the transmission lines are the edges in the power grid topology; in other words, the physical nodes are connected into a physical network by transmission lines. At the same time, the information network forms another dimension through optical transmission channels on this basis; in other words, there are communication lines between physical nodes, and the communication lines and nodes form an information network through communication relationships. The information network of physical nodes in the regional power grid consists of m directed graphs G1, G2,…, G m Indicates that, respectively, with RG1, RG2, ..., RG m Correspondingly, for the physical network node RG k , the communication within the regional power grid is represented by a directed graph G k =(V k ,E k ,A k ) construction, where the non-empty node set V k = {V k1 ,V k2 ,…,V km} represents a set of physical network nodes, node V ki With RG i Corresponding to the edge set Represents the communication lines between physical network nodes; A k =[a ij ] m×m ,i,j∈N RG , is the adjacency matrix, adjacency matrix A k The diagonal elements a ii = 0, if and only if RG i With RG j When there is a communication line between them, the off-diagonal element a ij >0,i≠j, that is, (V i ,V j )∈E k ; Node V i In-degree d in (V i ) is defined as Definition G k The in-degree matrix D k and the Laplace matrix L are D k =diag{d in (V i )} and L=D k -A k ;

[0042] Step S2: Design a self-healing grid control strategy based on droop control, combine the information-physical network, and introduce a predictive control strategy based on an event-triggered mechanism; the event-triggered predictive control strategy is carried out by the distributed energy storage node i on the information network within the regional power grid, and the virtual inertia correction signal is calculated through the state prediction consistency algorithm, which compensates for the frequency / voltage drop caused by the droop control in a distributed manner, and realizes the distribution of active power and reactive power among distributed energy storage nodes according to demand. Among them, "distributed energy storage node" includes "distributed energy storage equipment"; unless otherwise specified, the nodes in other parts of this article all represent "distributed energy storage nodes".

[0043] like Figure 1 As shown, the specific method of the self-healing control strategy of the droop control described in step S2 is:

[0044]

[0045] Among them, P0 is the rated active power, Q0 is the rated reactive power, P el is the node output active power, Q el Output reactive power for the node, is the state prediction control quantity based on event triggering of the node corresponding to the frequency in the power system, is the state prediction control quantity of the node corresponding to the voltage value in the power system based on event triggering, L is the system Laplace matrix, k pf is the active power droop coefficient, k qv is the reactive power droop coefficient, and C is the number of distributed energy storage nodes in the regional power grid.

[0046] Step S3: Design the communication strategy between distributed generators through event-triggered communication and design the event-triggered state predictive control algorithm.

[0047] like Figure 1 As shown, the specific method of the event-triggered state prediction control algorithm according to step S3 is:

[0048] For the i-th distributed energy storage device DE i , i∈N DE ,use represents the kth triggering moment of the ith node, Indicates the kth trigger interval. k represents the event triggering moment of the distributed energy storage system, and L represents the system Laplace matrix. Therefore, when the inequality of the following formula 3 is not satisfied, the event triggering executes the communication task and sends a reference signal to other distributed energy storage devices. The event triggering conditions are as follows:

[0049]

[0050] In the formula, ρ is the first constant parameter, ρ∈(0,1), γ is the second constant parameter, γ>0, ρ and γ ensure that there will be no infinite triggering behavior (Zeno effect) within the set minimum triggering time. i (t) represents the state of the i-th agent (distributed energy storage device) at the triggering moment With the current state ω i (t), in other words, e i (t) is the trigger condition. When the difference meets the condition, the control is triggered and applied. max (L), λ min (L) represents the maximum eigenvalue and minimum eigenvalue of the matrix L. ∥·∥ represents the Euclidean norm. z is the designed trigger condition parameter, through which the trigger rate can be adjusted while ensuring system stability.

[0051] On this basis, this method introduces the following state predictor:

[0052]

[0053] The self-healing control protocol with state predictor is as follows:

[0054]

[0055] Step S4: According to step S2 and step S3, combined with the event trigger prediction algorithm, the communication strategy between distributed energy storage nodes is optimized.

[0056] Step S5: Optimize the communication strategy of the distributed droop control algorithm to restore the output frequency and voltage of the distributed generator to the rated value while reducing the communication cost, that is, make the control error approach zero.

[0057] An embodiment is given below to illustrate the method of the present invention.

[0058] In this embodiment, a physical equivalent model of a regional power grid system is established in MATLAB / Simulink simulation software, such as Figure 3 As shown. The 10-machine 39-node standard system model is modified and verified. The system is an actual power network, including 39 nodes, 34 lines, including 10 power plant nodes. The system base power is 100MVA and the base voltage is 345kV. It is assumed that each power plant node is equipped with a 10% / 1h energy storage system according to capacity, and the power generation equipment itself does not participate in the regulation.

[0059] Table 1 RG node parameters and initial values

[0060]

[0061]

[0062] Table 2DE node parameters and initial values

[0063]

[0064]

[0065] Table 3 Line parameters between physical network nodes RG

[0066]

[0067]

[0068] The simulation process is set up as follows:

[0069] Simulation experiment 1: When the system is running at t=0s, a three-phase short circuit fault occurs on line 14-15. Due to the action of relay protection, the short circuit fault line is cut off at t=0.2s.

[0070] Simulation experiment 2: Under the same fault conditions as experiment 1, a controller with conventional droop control protocol is added to the system.

[0071] Simulation experiment three: Under the same fault conditions as experiment one, a controller based on event-triggered optimization droop control protocol and a controller containing event-triggered state prediction mechanism are added to the system.

[0072] When a partial disconnection occurs in the system, the frequency / voltage balance effect after system power flow operation is as follows: Figure 4 As shown, this is a schematic diagram of the DE node export frequency. Figure 5 Schematic diagram of the DE node export voltage. Due to the transient characteristics of the regional power grid system, the frequency / voltage of DE will tend to a new dynamic stable state in a long time after a certain fluctuation.

[0073] Under the same fault conditions in the experiment, the frequency / voltage regulation effect after adding the controller of the conventional droop control protocol in the DE node is as follows: Figure 6 , Figure 7 As shown in the figure, it can be seen that at t=8s, after the adjustment of conventional droop control, the entire regional power grid tends to be stable.

[0074] Under the same fault conditions in the experiment, after adding the controller based on the event-triggered optimization droop control protocol and the controller containing the event-triggered state prediction mechanism to the system, the frequency / voltage balance effect after system power flow operation is as follows Figure 8 , Fig. 9As shown, at t=5.5s, after adjustment by the self-healing controller of the present method, the entire regional power grid can tend to a stable state more quickly and recover to the rated operating state.

[0075] In Experiment 3, the time from the occurrence of a fault to the restoration of a stable state is compared with the communication time generated by the control demand, and multiple experiments are performed to obtain the average value. The experiment shows that the event trigger with state prediction can effectively reduce the communication demand by more than 50%, such as Fig.10 As shown, this saves communication and computing resource consumption while ensuring the system self-healing control requirements.

[0076] It can be seen from this embodiment that the method proposed in the present invention can effectively realize the distributed energy storage scheduling in the regional power grid to achieve strong support control for the self-healing regulation of the system. On the basis of maintaining stability based on the droop control of event-triggered optimization, the frequency and voltage of the system nodes can be restored to the rated value. The designed control algorithm has good convergence performance. The communication time and bandwidth between DE nodes are greatly reduced, which significantly reduces the operating cost of the entire system.

[0077] The above description is only a preferred embodiment of the present invention and is not intended to be a further limitation of the present invention. All equivalent changes made using the contents of the present specification and drawings are within the protection scope of the present invention.

Claims

1. A method for controlling a regional distributed energy storage self-healing power grid in a power system, characterized in that: The following steps are involved: Step S1: Establishing an information-physical network corresponding to the regional power grid system, including a physical network and an information network, wherein the physical network provides power transmission and the information network provides information interaction; the information network includes a regional power grid intra-communication network composed of energy storage nodes of the regional power grid system; Step S2: Design a self-healing power grid control strategy based on droop control, combine the information-physical network, and introduce a predictive control strategy based on an event-triggered mechanism; the event-triggered predictive control strategy exchanges information on the information network within the regional power grid of the distributed energy storage node, calculates the virtual inertia correction signal through the state prediction consistency algorithm, and compensates for the frequency / voltage drop caused by the droop control in a distributed manner, and realizes the distribution of active power and reactive power among the distributed energy storage nodes according to demand; Step S3: Designing an event-triggered state predictive control algorithm; Step S4: Optimizing the communication strategy between distributed energy storage nodes through an event-triggered state predictive control algorithm; Step S5: Optimize the communication strategy of the distributed droop control algorithm to restore the output frequency and voltage of the distributed generator to the rated value while reducing the communication cost, that is, make the control error approach zero.

2. The method for controlling a regional distributed energy storage self-healing power grid in a power system according to claim 1, characterized in that: In the information-physical network, the frequency and voltage of physical nodes are dynamically correlated, and different nodes play different functional roles in the physical network. By abstracting the physical network into a complex network with multiple nodes and multiple lines, the nodes include at least one of power plants, substations, energy storage, and loads, and the transmission line branches are the edges in the power grid topology. At the same time, the information network constitutes another dimension on this basis.

3. The method for controlling a regional distributed energy storage self-healing power grid in a power system according to claim 2, characterized in that: In the droop control loop, power scheduling is achieved by changing the power-frequency droop coefficient. When the power command issued is P el / Q el When the power is adjusted, the power dispatching and frequency deviation reduction are realized by changing the droop power curve; the droop coefficient after adjustment is: Among them, P0 is the rated active power, Q0 is the rated reactive power, P el Node output active power, Q el Output reactive power for the node, is the state prediction control quantity based on event triggering of the node corresponding to the frequency in the power system, is the state prediction control quantity of the node corresponding to the voltage value in the power system based on event triggering, L is the system Laplace matrix, k pf is the active power droop coefficient, k qv is the reactive power droop coefficient, and C is the number of distributed energy storage nodes in the regional power grid.

4. The method for controlling a regional distributed energy storage self-healing power grid in a power system according to claim 3, characterized in that: For the i-th distributed energy storage device DE i , i∈N DE , N DE As a collection of energy storage devices, represents the kth triggering moment of the ith node, Indicates the kth trigger interval; r k represents the event triggering moment of the distributed energy storage system, and L represents the system Laplace matrix; therefore, when the inequality of the following formula 3 is not satisfied, the event triggering executes the communication task and sends a reference signal to other distributed energy storage devices; the event triggering conditions are as follows: Where ρ is the first constant parameter, ρ∈(0,1), γ is the second constant parameter, γ>0, e i (t) represents the state of the i-th agent at the triggering moment With the current state ω i The difference of (t), λ max (L), λ min (L) represents the maximum eigenvalue and minimum eigenvalue of the matrix L, respectively, and z is the trigger condition parameter; On this basis, the following state predictor is introduced: The self-healing control protocol with state predictor is as follows: