A control system and method for distributed energy storage participating in power system frequency regulation

Abstract

The present invention discloses a control system and method for distributed energy storage to participate in power system frequency regulation, which belongs to the field of energy storage cluster control. It includes: a distributed state of charge balancing control system within the energy storage cluster, which is used for the energy storage controllers within the same energy storage cluster to communicate with each other and generate a reduced-order aggregation model of the energy storage cluster for uploading; after receiving the frequency modulation control signal, the energy storage controllers within the same energy storage cluster communicate with each other, with the energy storage state of charge balancing as the goal, distribute the frequency modulation control signal, and send the distribution result to each energy storage; an energy storage cluster frequency regulation system, which is used to receive the reduced-order aggregation model and establish a prediction model for energy storage to participate in frequency modulation; a cost function is constructed with the received power system frequency offset as the independent variable, and the prediction model of energy storage participating in frequency modulation is used as the constraint to solve the cost function and obtain the frequency modulation control signal of each energy storage cluster for sending. The present invention can reduce the order of the frequency modulation optimization problem, reduce the computational burden, and promote the complementarity of each energy storage within the cluster.

Description

Technical Field

[0001] The present invention belongs to the technical field of distributed energy storage control, and more specifically, relates to a control system and method for distributed energy storage participating in power system frequency regulation. Background Art

[0002] The intermittent and random nature of growing renewable energy capacity exacerbates the problem of uncertain, instantaneous power fluctuations in the power system, posing a challenge to power system frequency regulation. Due to ramp rate limitations, traditional generators struggle to handle frequency excursions caused by power fluctuations. Energy storage, with its rapid installation and flexible response, is an ideal option for assisting power system frequency regulation.

[0003] Large amounts of small-capacity distributed energy storage play a vital role in grid frequency regulation. However, without a suitable control solution, this can lead to increased computational burdens and inefficient utilization. For centralized control, a large number of controlled objects leads to a significant optimization computational burden and heavy communication burdens between the central controller and the controlled devices. For distributed control, a large amount of energy storage leads to complex sparse communication networks and slow convergence. Summary of the Invention

[0004] In response to the shortcomings of existing technologies and the need for improvement, the present invention provides a control system and method for distributed energy storage participating in power system frequency regulation. The goal is to apply reduced-order models to frequency control to reduce the system order. The lower layer only requires communication between adjacent energy storage units, completing state-of-charge balancing control within the energy storage cluster within a distributed control framework. The upper layer uses the reduced-order model provided by the lower layer to perform frequency regulation and sends the generated frequency control signal to the energy storage cluster, significantly reducing the computational burden. Using this control scheme, frequency regulation and energy storage state-of-charge balancing can be comprehensively considered.

[0005] To achieve the above objectives, the present invention provides a control system for distributed energy storage participating in power system frequency regulation, comprising:

[0006] An inter-energy storage cluster frequency regulation system, multiple energy storage cluster state-of-charge balancing control systems corresponding one-to-one to each energy storage cluster, and each energy storage cluster state-of-charge balancing control system includes multiple energy storage controllers corresponding one-to-one to each energy storage;

[0007] The state of charge balancing control system within the energy storage cluster is used to generate a reduced-order aggregation model of the energy storage cluster through communication between the energy storage controllers within the same energy storage cluster, and upload it to the frequency regulation system between energy storage clusters. After receiving the frequency modulation control signal, the energy storage controllers within the same energy storage cluster communicate with each other, distribute the frequency modulation control signal with the goal of balancing the state of charge of the entire energy storage cluster, and send the distribution result to each energy storage;

[0008] The frequency regulation system between energy storage clusters is used to receive the reduced-order aggregation model sent by each energy storage cluster and establish a prediction model for energy storage's participation in frequency regulation. It constructs a cost function using the received power system frequency offset, frequency change rate, and control cost as independent variables, and solves the cost function using the prediction model of energy storage's participation in frequency regulation as a constraint to obtain the frequency regulation control signal for each energy storage cluster and send it to the charge state balancing control system within the energy storage cluster.

[0009] Preferably, the prediction model for energy storage participating in frequency regulation is as follows:

[0010]

[0011] in,

[0012] Where t represents the discrete time step, X k (t) represents the state variable sequence of the power system in the kth zone at the tth step, X l (t) represents the state variable sequence of the power system in area l at the tth step, U k (t) represents the control variable sequence of the power system in the kth area at the tth step, Y k (t) represents the output variable sequence of the power system in the kth zone at the tth step, S A,kk represents the system matrix of the k-th region power system with respect to local variables, S B represents the control matrix of the power system, S A,kl represents the system matrix of the kth region power system with respect to the lth region power system variables, S C represents the output matrix of the power system, x k (t) represents the state variable of the power system in the kth zone at the tth step, N p represents the prediction time step, u k (t) represents the control variable of the power system in the kth zone at the tth step, N c represents the control time step, y k (t) represents the output variable of the power system in the kth zone at the tth step.

[0013] Beneficial effects: In order to solve the problem that existing proportional-integral control and adaptive control are difficult to add constraints, the present invention establishes a prediction model to predict the future output of the system based on the historical information of the control object and future inputs. Since the constraints on the output variables can be converted into constraints on the input variables through mathematical operations, effective constraints on the output variables are achieved.

[0014] Preferably, the cost function is:

[0015]

[0016] Where k represents the region number, q represents the weight related to the frequency offset, and N p represents the prediction step length, ω k (t) represents the frequency offset of the power system in the kth zone at the tth step, r represents the weight related to the control variable, N c represents the control step size, represents the control variable at the kth step, α represents the weight related to the frequency offset, and β represents the weight related to the frequency change rate per unit time.

[0017] Beneficial Effects: If only the frequency offset is used as the state variable of the cost function, the frequency change rate will be too high, and when the frequency control is completed, the control variable cannot return to 0. To address this problem, the present invention incorporates the frequency change rate and the state variables of the control variable. Due to the optimized calculation result, the cost function is minimized, achieving the purpose of simultaneously suppressing the frequency offset and the frequency change rate. When the other terms of the cost function are 0, the control variable is also guaranteed to be 0.

[0018] Preferably, the optimal control sequence U is obtained k T (t)=[u k T (t),u k T (t+1)…,u k T (t+N c -1)], take the first item u k T (t) = P * k , sent to the corresponding k-th energy storage cluster;

[0019] Among them, U k (t) represents the control signal sequence of the power system in the kth zone at the tth step, u k T (t) represents the control signal of the power system in the kth zone at the tth step, u k T (t) = P * k Represents the frequency modulation control signal of the kth energy storage cluster.

[0020] Beneficial effects: To address the problem of limited accuracy of the prediction model, the present invention performs rolling optimization by taking only the first item of the optimal control sequence as the control signal. Since the error of each forward step will gradually accumulate, only the prediction error of the forward step is the smallest, thereby achieving the purpose of minimizing the control error.

[0021] Preferably, based on the frequency response model of the grid-type energy storage, a reduced-order aggregation model of the energy storage cluster is obtained by weighted averaging according to the energy storage capacity.

[0022] Beneficial effect: In order to address the problem of inconsistent with engineering practice caused by existing control strategies that do not consider the system model, the present invention takes into account the energy storage frequency response model. Since the power response of the energy storage itself has a time delay and the controller also has a corresponding time delay, a modeling that is closer to engineering practice is achieved.

[0023] Preferably, the goal of balancing the state of charge of the entire energy storage cluster is specifically as follows:

[0024]

[0025] Among them, SoC i represents the state-of-charge of the i-th energy storage, P i represents the output active power of the i-th energy storage, n k Represents the amount of energy storage in the kth energy storage cluster.

[0026] Beneficial effect: In order to solve the problem of overcharging and over-discharging of part of the energy storage in the system due to the existing power distribution method not considering the charge state balance, the present invention uses charge state balance control. During discharge, the energy storage with a high charge state can compensate for the power of the energy storage with a low charge state, and during charging, the energy storage with a low charge state can compensate for the power of the energy storage with a high charge state, thereby avoiding the problem of overcharging and over-discharging of part of the energy storage.

[0027] Preferably, within the energy storage cluster, master-slave consistency control is used to achieve charge state balance control of the entire energy storage cluster. The active power command value of the energy storage as the leader (marked as 1) is:

[0028]

[0029] Among them, P ref1 Indicates the main energy storage active power command value, represents the active power command sent by the inter-storage cluster frequency regulation system to the kth energy storage cluster, n k Represents the amount of energy storage in the kth energy storage cluster.

[0030] The remaining energy storage is the follower, and its active power command value is updated according to the following formula:

[0031]

[0032]

[0033] Among them, P refiIndicates the active power command value of the i-th slave energy storage, u i Indicates the update rate of the active power command value of the i-th slave energy storage, P i and P j Represent the output active power of the i-th and j-th energy storage, γ i and γ j represents the local control variable of the i-th and j-th energy storage, which is defined as:

[0034]

[0035] Among them, P i represents the output active power of the i-th energy storage, F soci represents the function related to the i-th energy storage state of charge, which is defined as:

[0036]

[0037] Among them, P i represents the output active power of the i-th energy storage, SoC min and SoC max are the lower and upper limits of the energy storage state of charge, respectively.

[0038] Beneficial effect: To address the problem of power circulation and increased transmission loss caused by existing state of charge balancing control, the present invention proposes local control variables for the state of charge, so that all energy storage in the same cluster can be charged or discharged at the same time, avoiding the problem of power circulation.

[0039] The present invention also provides a control method for distributed energy storage participating in power system frequency regulation, comprising the following steps:

[0040] The energy storage controllers within the same energy storage cluster communicate with each other to generate a reduced-order aggregation model for the energy storage cluster. After receiving the frequency modulation control signal, the energy storage controllers within the same energy storage cluster communicate with each other to distribute the frequency modulation control signal with the goal of balancing the state of charge of the entire energy storage cluster, and send the distribution results to each energy storage.

[0041] After receiving the reduced-order aggregation model sent by each energy storage cluster, a prediction model for energy storage participating in frequency regulation is established; a cost function is constructed with the received power system frequency offset, frequency change rate, and control cost as independent variables, and the prediction model for energy storage participating in frequency regulation is used as a constraint to solve the cost function and obtain the frequency regulation control signal of each energy storage cluster.

[0042] In general, the above technical solutions conceived by the present invention can achieve the following beneficial effects:

[0043] The existing frequency regulation power sources in the power system are mainly thermal power units, which have long response time lags, low unit ramp-up rates, and problems such as regulation delays, regulation deviations, and regulation reversals. The energy storage system has a fast response speed, strong short-term power throughput, and flexible regulation. The combination of energy storage systems and thermal power units can effectively improve the frequency regulation capabilities of the power system. However, a region usually contains many energy storage units, which have different capacities, response speeds, economic indicators, and dispersed layouts. If they are allowed to be distributed and connected to the power grid, problems such as dispersed output and difficulty in regulation will arise. Only by aggregating them into an energy storage group for unified regulation can a synergistic effect of energy storage scale be formed, energy utilization efficiency can be improved, and regulation can be facilitated. The present invention proposes a control scheme for distributed energy storage to participate in the frequency regulation of the power system by establishing a reduced-order aggregation model of the energy storage cluster. Since the order of the frequency regulation optimization problem is reduced, the computational burden is greatly reduced and the complementary effects of the energy storage within the cluster are promoted. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 is the topology of the power system under test;

[0045] Figure 2 Schematic diagrams of distributed energy storage participating in power system frequency regulation. (a) is a schematic diagram of grid-connected energy storage participating in power system frequency regulation. (b) is a schematic diagram of synchronous motors participating in power system frequency regulation.

[0046] Figure 3 This is a schematic diagram of the control system for distributed energy storage participating in power system frequency regulation;

[0047] Figure 4 This is a schematic diagram of master-slave consistency control within the k-th energy storage cluster;

[0048] Figure 5 Figure 1 is the control effect diagram of each energy storage cluster, (a) is the frequency curve under different control weights, (b) is the frequency change rate curve under different control weights, (c) is the active power output curve of each energy storage in cluster 1, and (d) is the SoC curve of each energy storage in cluster 1. DETAILED DESCRIPTION

[0049] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely for the purpose of explaining the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.

[0050] The present invention provides a control system for distributed energy storage participating in power system frequency regulation, comprising:

[0051] An inter-energy storage cluster frequency regulation system, multiple energy storage cluster state-of-charge balancing control systems corresponding one-to-one to each energy storage cluster, and each energy storage cluster state-of-charge balancing control system includes multiple energy storage controllers corresponding one-to-one to each energy storage;

[0052] The state of charge balancing control system within the energy storage cluster is used to generate a reduced-order aggregation model of the energy storage cluster through communication between the energy storage controllers within the same energy storage cluster, and upload it to the frequency regulation system between energy storage clusters. After receiving the frequency modulation control signal, the energy storage controllers within the same energy storage cluster communicate with each other, distribute the frequency modulation control signal with the goal of balancing the state of charge of the entire energy storage cluster, and send the distribution result to each energy storage;

[0053] The frequency regulation system between energy storage clusters is used to receive the reduced-order aggregation model sent by each energy storage cluster and establish a prediction model for energy storage's participation in frequency regulation. It constructs a cost function using the received power system frequency offset, frequency change rate, and control cost as independent variables, and solves the cost function using the prediction model of energy storage's participation in frequency regulation as a constraint to obtain the frequency regulation control signal for each energy storage cluster and send it to the charge state balancing control system within the energy storage cluster.

[0054] Figure 1 A scenario where a synchronous motor and an energy storage cluster jointly participate in system frequency regulation is used as an example.

[0055] like Figure 2 As shown in Figure 1, energy storage participates in system frequency regulation through droop control, and synchronous motor participates in system frequency regulation through primary and secondary frequency regulation.

[0056] like Figure 3 As shown in the figure, based on the frequency response model of energy storage, the reduced-order aggregation model of the energy storage cluster is obtained by weighted average of energy storage capacity:

[0057]

[0058] where ω k represents the frequency offset of the kth energy storage cluster, P k represents the output active power of the kth energy storage cluster, P * k represents the active power command of the kth energy storage cluster, T k and R pk is the equivalent delay and equivalent droop coefficient of the low-order aggregation model of the k-th energy storage cluster.

[0059]

[0060] Among them, S i is the rated capacity of energy storage i, T i is the inertia coefficient of energy storage i, T i and R piis the delay and droop coefficient of the energy storage i.

[0061] Based on the energy storage reduced-order aggregation model, a prediction model for energy storage participation in frequency regulation is established. Using frequency offset, frequency change rate and control cost as the cost function, the optimal control sequence is solved and sent to each energy storage cluster.

[0062] Based on the reduced-order aggregation model, a prediction model is established as follows:

[0063]

[0064] Among them, x k and u k represents the local state variables and control variables of the k-th energy storage cluster, x l is the state variable vector of cluster l, y k is the output variable vector of cluster k, A kk is the state matrix associated with the local state variables, A kl is the state matrix related to the state variables of cluster l, B is the control matrix, and C is the output matrix.

[0065] In the embodiment,

[0066]

[0067]

[0068] After discretization, we get:

[0069]

[0070] Among them, x k (t) is the state variable of the k-th energy storage cluster at the t-th step, u k (t) is the control variable of the k-th energy storage cluster at the t-th step, y k (t) is the output variable vector of the k-th energy storage cluster at the t-th step. The discrete state space matrices are expressed as:

[0071]

[0072]

[0073]

[0074] C d =C

[0075] Among them, t d is the discretized sampling time.

[0076] According to the discretized state space equation, the control step size is N c, system future N p The state prediction equation within the step is:

[0077]

[0078] in,

[0079]

[0080]

[0081]

[0082]

[0083]

[0084] Among them, A d,kk A represents the discrete state matrix of the k-th energy storage cluster with respect to local variables. d,kl represents the discrete state matrix of the kth energy storage cluster with respect to the lth energy storage cluster variable, B d represents the discrete control matrix, C d represents the discrete output matrix.

[0085] The cost function is:

[0086]

[0087] The cost function of the kth energy storage cluster is related to that of the other clusters, indicating that frequency control at the upper level is a coupled optimization problem. Therefore, only through information exchange and iteration among the controllers of different clusters can the global optimal solution be achieved. The iterative rules of the distributed optimization method are described below.

[0088] 1) The initial state x of all clusters k (num) =0, the number of iterations num is recorded as 1.

[0089] 2) Solve the optimization problem and obtain the optimal control signal

[0090] 3) By setting x k (num) and Substitute into the state equation to calculate the state x k (num+1) .

[0091] 4) Exchange information with other clusters through communication links, and then obtain the optimal control signal by solving the optimization problem

[0092] 5) If (where ε is a small positive value), the condition for ending the iteration for cluster k is met. If this condition is met in all clusters, the system has reached equilibrium, is the global optimal solution for group k. Otherwise, num=num+1, and return to step (3).

[0093] Get the optimal control sequence U k T (t)=[u k T (t),u k T (t+1)…,u k T (t+N c -1)], take the first item u k T (t) = P * k , sent to the corresponding k-th energy storage cluster until the end of the control cycle.

[0094] Within the energy storage cluster, the master-slave consensus algorithm is used to distribute the control signal among the energy storage units in the cluster, such as Figure 4 shown.

[0095] As the leader energy storage (marked as 1), its active power command value is

[0096]

[0097] Among them, P ref1 Indicates the main energy storage active power command value, represents the active power command sent by the inter-storage cluster frequency regulation system to the kth energy storage cluster, n k Represents the amount of energy storage in the kth energy storage cluster.

[0098] The remaining energy storage is the follower, and its active power command value is updated according to the following formula:

[0099]

[0100]

[0101] Among them, P refi Indicates the active power command value of the i-th slave energy storage, u i Indicates the update rate of the active power command value of the i-th slave energy storage, P i and P j Represent the output active power of the i-th and j-th energy storage, γ i and γj represents the local control variable of the i-th and j-th energy storage, which is defined as:

[0102]

[0103] Among them, P i represents the output active power of the i-th energy storage, F soci represents the function related to the i-th energy storage state of charge, which is defined as:

[0104]

[0105] Among them, P i represents the output active power of the i-th energy storage, SoC min and SoC max are the lower and upper limits of the energy storage state of charge, respectively.

[0106] In the present invention, the upper layer distributes the secondary frequency modulation signal among clusters based on the reduced-order aggregation model of the energy storage cluster. The lower layer provides an aggregation reduced-order model of the energy storage cluster and performs charge state balancing control within the cluster for the control signal provided by the upper layer. Frequency regulation and charge state balancing are comprehensively considered, avoiding the problem of selecting weight coefficients of multi-objective cost functions; effectively reducing the dimensionality of the prediction model, alleviating the computational and communication burden of the upper-layer frequency control, and being suitable for controlling a large number of small-capacity and multi-type energy storage.

[0107] In this embodiment, a 150MW load step change is set and the system response is observed to illustrate the applicability and superiority of the control strategy proposed in the present invention.

[0108] Figure 5 Figures (a) and (b) compare the system frequency deviation and ROCOF under different weighting coefficients. When α = 1 and β = 0, the control signal effectively suppresses frequency deviation, but this results in frequency overshoot and a correspondingly large ROCOF. Conversely, when α = 0 and β = 1, the ROCOF is effectively reduced, but a frequency deviation of nearly 41.7% is generated, and frequency overshoot occurs. Therefore, the weighting coefficients should be carefully selected to balance frequency deviation and ROCOF. After comparing different coefficients, α = 0.3 and β = 0.7 were selected.

[0109] Figure 5 (c) in Figure 1 shows the output power of each energy storage in cluster 1. The control signal is decomposed and sent to each energy storage through the master-slave consensus algorithm. Figure 5 (d) in the figure shows the state of charge of each energy storage in cluster 1, which indicates that the energy storage with a higher state of charge is discharged more to balance the state of charge of each energy storage and avoid overcharging or over-discharging.

[0110] The present invention also provides a control method for distributed energy storage participating in power system frequency regulation, comprising the following steps:

[0111] The energy storage controllers within the same energy storage cluster communicate with each other to generate a reduced-order aggregation model for the energy storage cluster. After receiving the frequency modulation control signal, the energy storage controllers within the same energy storage cluster communicate with each other to distribute the frequency modulation control signal with the goal of balancing the state of charge of the entire energy storage cluster, and send the distribution results to each energy storage.

[0112] After receiving the reduced-order aggregation model sent by each energy storage cluster, a prediction model for energy storage participating in frequency regulation is established; a cost function is constructed with the received power system frequency offset, frequency change rate, and control cost as independent variables, and the prediction model for energy storage participating in frequency regulation is used as a constraint to solve the cost function and obtain the frequency regulation control signal of each energy storage cluster.

[0113] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code. The scheme in the embodiment of the present application can be implemented in various computer languages, for example, object-oriented programming language Java and literal translation scripting language JavaScript, etc.

[0114] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the steps in the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0115] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1The function specified in one or more boxes.

[0116] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 The steps for the function specified in one or more boxes.

[0117] Although preferred embodiments of the present application have been described, those skilled in the art may make additional changes and modifications to these embodiments once they are aware of the basic inventive concepts. It is clear that those skilled in the art may make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, to the extent such changes and modifications fall within the scope of the present application and its equivalents, the present application is intended to encompass such changes and modifications.

Claims

1. A control system for distributed energy storage participating in power system frequency regulation, characterized in that: include: An inter-energy storage cluster frequency regulation system, multiple energy storage cluster state-of-charge balancing control systems corresponding one-to-one to each energy storage cluster, and each energy storage cluster state-of-charge balancing control system includes multiple energy storage controllers corresponding one-to-one to each energy storage; The energy storage cluster internal state of charge balancing control system is configured to generate a reduced-order aggregation model of the energy storage cluster through mutual communication between energy storage controllers within the same energy storage cluster, and upload the model to the inter-energy storage cluster frequency regulation system. Upon receiving a frequency modulation control signal, the energy storage controllers within the same energy storage cluster communicate with each other, distribute the frequency modulation control signal with the goal of balancing the state of charge of the entire energy storage cluster, and send the distribution result to each energy storage. Within the energy storage cluster, master-slave consistency control is used to achieve charge balancing control of the entire energy storage cluster, and the active power command value of the master energy storage is: in, P ref1 Indicates the main energy storage active power command value, represents the active power instruction sent by the inter-storage cluster frequency regulation system to the kth energy storage cluster, n k Indicates the k The amount of energy storage in each energy storage cluster; The active power command value from the energy storage is updated according to the following formula: in, P refi Indicates the i The active power command value from the energy storage, u i Indicates the i The update rate of the active power command value from the energy storage, P i and P j Respectively represent i and j The output active power of each energy storage, γ i and γ j Indicates the i and j The local control variable of the energy storage is defined as: in P i Indicates the i The output active power of each energy storage, F soci Indicates the i The function related to the state of charge of the energy storage is defined as: SoC i Indicates the i The state of charge of a stored energy, SoC min and SoC max are the lower and upper limits of the energy storage state of charge, respectively; The frequency regulation system between energy storage clusters is used to receive the reduced-order aggregation model sent by each energy storage cluster and establish a prediction model for energy storage's participation in frequency regulation. It constructs a cost function using the received power system frequency offset, frequency change rate, and control cost as independent variables, and solves the cost function using the prediction model of energy storage's participation in frequency regulation as a constraint to obtain the frequency regulation control signal for each energy storage cluster and send it to the charge state balancing control system within the energy storage cluster.

2. The control system according to claim 1, wherein: The prediction model for energy storage participating in frequency regulation is as follows: in, t represents the discrete time step, X k ( t ) indicates the t Step 1 k The state variable sequence of the regional power system, X l ( t ) indicates the t Step 1 l The state variable sequence of the regional power system, U k ( t ) indicates the t Step 1 k The control variable sequence of the district power system, Y k ( t ) indicates the t Step 1 k The output variable sequence of the regional power system, Indicates the k The system matrix of the district power system with respect to local variables, represents the control matrix of the power system, Indicates the k District power system about l The system matrix of regional power system variables, represents the output matrix of the power system, x k ( t ) indicates the k District Power System t The state variables at the time of step, represents the prediction time step, u k ( t ) indicates the k District Power System t The control variable at step time, represents the control time step, y k ( t ) indicates the k District Power System t Output variable at step time.

3. The control system according to claim 2, wherein: The cost function is: in, Indicates the area number, q represents the weight associated with the frequency offset, represents the prediction step length, ω k ( t ) indicates the k District Power System t Step frequency offset, r represents the weight associated with the control variable, represents the control step length, represents the control variable at the kth step, α represents the weight associated with the frequency offset, β Represents the weight related to the rate of change of frequency per unit time; Get the optimal control sequence U k T ( t )=[ u k T ( t ), u k T ( t +1)…, u k T ( t + N c -1)], take the first item u k T ( t ) = P * k , sent to the corresponding k energy storage clusters; in, U k ( t ) indicates the k District Power System t The control signal sequence of the step, u k T ( t ) indicates the k District Power System t The control signal of the step, u k T ( t ) = P * k Indicates the k The frequency modulation control signal of the energy storage cluster.

4. The control system according to claim 1, wherein: The goal of balancing the state of charge of the entire energy storage cluster is as follows: 。 5. A control method for distributed energy storage participating in power system frequency regulation, characterized in that: The following steps are involved: The energy storage controllers within the same energy storage cluster communicate with each other to generate a reduced-order aggregation model for the energy storage cluster. After receiving the frequency modulation control signal, the energy storage controllers within the same energy storage cluster communicate with each other, distribute the frequency modulation control signal with the goal of balancing the charge state of the entire energy storage cluster, and send the distribution results to each energy storage. Within the energy storage cluster, master-slave consistency control is used to achieve charge balance control for the entire energy storage cluster. The active power command value of the master energy storage is: in, P ref1 Indicates the main energy storage active power command value, represents the active power instruction sent by the inter-storage cluster frequency regulation system to the kth energy storage cluster, n k Indicates the k The amount of energy stored in each energy storage cluster; The active power command value from the energy storage is updated according to the following formula: in, P refi Indicates the i The active power command value from the energy storage, u i Indicates the i The update rate of the active power command value from the energy storage, P i and P j Respectively represent i and j The output active power of each energy storage, γ i and γ j Indicates the i and j The local control variable of the energy storage is defined as: in P i Indicates the i The output active power of each energy storage, F soci Indicates the i The function related to the state of charge of the energy storage is defined as: SoC i Indicates the i The state of charge of a stored energy, SoC min and SoC max are the lower and upper limits of the energy storage state of charge, respectively; After receiving the reduced-order aggregation model sent by each energy storage cluster, a prediction model for energy storage participating in frequency regulation is established; a cost function is constructed with the received power system frequency offset, frequency change rate, and control cost as independent variables, and the prediction model for energy storage participating in frequency regulation is used as a constraint to solve the cost function and obtain the frequency regulation control signal of each energy storage cluster.

6. The control method according to claim 5, wherein: The prediction model for energy storage participating in frequency regulation is as follows: in, t represents the discrete time step, X k ( t ) indicates the t Step 1 k The state variable sequence of the regional power system, X l ( t ) indicates the t Step 1 l The state variable sequence of the regional power system, U k ( t ) indicates the t Step 1 k The control variable sequence of the district power system, Y k ( t ) indicates the t Step 1 k The output variable sequence of the regional power system, Indicates the k The system matrix of the district power system with respect to local variables, represents the control matrix of the power system, Indicates the k District power system about l The system matrix of regional power system variables, represents the output matrix of the power system, x k ( t ) indicates the k District Power System t The state variables at the time of step, represents the prediction time step, u k ( t ) indicates the k District Power System t The control variable at step time, represents the control time step, y k ( t ) indicates the k District Power System t Output variable at step time.

7. The control method according to claim 6, wherein: The cost function is: in, Indicates the area number, q represents the weight associated with the frequency offset, represents the prediction step length, ω k ( t ) indicates the k District Power System t Step frequency offset, r represents the weight associated with the control variable, represents the control step length, represents the control variable at the kth step, α represents the weight associated with the frequency offset, β Represents the weight related to the rate of change of frequency per unit time; Get the optimal control sequence U k T ( t )=[ u k T ( t ), u k T ( t +1)…, u k T ( t + N c -1)], take the first item u k T ( t ) = P * k , sent to the corresponding k energy storage clusters; in, U k ( t ) indicates the k District Power System t The control signal sequence of the step, u k T ( t ) indicates the k District Power System t The control signal of the step, u k T ( t ) = P * k Indicates the k The frequency modulation control signal of the energy storage cluster.

8. The control method according to claim 5, wherein: The goal of balancing the state of charge of the entire energy storage cluster is as follows: 。

Images