Data-driven distributed voltage control method and device for distributed power supply and energy storage

Abstract

The application provides a data-driven distributed power supply and energy storage voltage control method and device, and relates to the technical field of distribution network voltage regulation.The method comprises the following steps: obtaining power grid operation data and power grid operation parameters, and initializing a control parameter k=1 and a control time t=0;obtaining node voltage data based on the power grid operation data and the power grid operation parameters;obtaining a voltage deviation threshold value, and determining whether the voltage deviation between the node voltage data and a node voltage reference value exceeds the voltage deviation threshold value;if yes, obtaining node injection power measurement data based on the power grid operation data;and if no, setting the control time t=t+Δt.The method provided by the application can alleviate the technical problem that the voltage regulation mode in the prior art is affected by the slow discrete regulation mode and cannot effectively cope with the voltage fluctuation caused by the output uncertainty of the distributed power supply.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of grid voltage control, and in particular to a data-driven distributed power source and energy storage voltage control method and device. BACKGROUND

[0002] With the high penetration of renewable energy in distribution networks, the power quality and reliability of distribution networks are negatively affected. Traditional regulation methods such as on-load voltage regulating transformers and capacitor banks, although can alleviate voltage out-of-limit to some extent, are affected by their discrete slow regulation mode and cannot effectively respond to voltage fluctuations caused by the output uncertainty of distributed power sources. Since distributed power sources only produce rated active power at some time, the remaining capacity of the inverter can be used for continuous voltage regulation, so the reactive power control of distributed power sources is a way to achieve fast voltage regulation. At the same time, distributed energy storage is widely connected to distribution networks due to its convenience and flexibility. The wide application of distributed energy storage provides fast response capacity for distribution networks, helps to smooth voltage fluctuations and regulate energy time and space distribution, effectively improving the controllability and flexibility of distribution network operation. However, without coordination means, the voltage regulation effect of the distributed energy storage optimization strategy and the reactive power control strategy of the distributed power source may conflict, leading to voltage deterioration and negatively affecting the safe and stable operation of the distribution network. Therefore, coordinated control of distributed power sources and distributed energy storage is of great significance to ensure the safe and economic operation of distribution networks.

[0003] In addition, in actual complex operating environments, the accurate parameters of distribution networks are difficult to obtain. The rapid development of smart measurement terminals and communication networks has promoted the high informatization of distribution and the development of data-driven control methods. Data-driven does not rely on detailed mathematical model information of the controlled, but only uses measurement data to statistically describe the input-output relationship of complex links to simulate and build unknown characteristics of complex links. Data-driven control can effectively solve the problem of lack of accurate parameters in modeling, effectively avoid the limitations of traditional mechanism modeling, and has strong applicability to highly nonlinear and strong uncertainty.

[0004] The distributed power source and energy storage coordinated voltage control method combines the data-driven distribution network voltage control method and the predictive control method, fully considers the time sequence characteristics of distributed energy storage and the mutual influence of distributed power sources and energy storage in the control process. This method does not require detailed mathematical models of distribution networks, but only needs to establish a data model by dynamically describing the input-output relationship of distribution networks according to the operation data of distribution networks, and then establish a coordinated control model of distributed power sources and energy storage, which can effectively realize voltage control of distribution networks. SUMMARY

[0005] Therefore, the present application aims to provide a data-driven distributed power supply and energy storage voltage control method and device, which can effectively deal with the voltage fluctuation caused by the output uncertainty of the distributed power supply by dynamically depicting the input-output relationship of the distribution network, establishing a data model, and then establishing a coordinated control model of the distributed power supply and energy storage, effectively realizing the voltage control of the distribution network, and improving the safety of the safe operation of the power grid.

[0006] In a first aspect, the present application provides a data-driven distributed power supply and energy storage voltage control method, which specifically includes the following steps:

[0007] Obtain power grid operation data and power grid operation parameters, including distributed energy storage type, distributed energy storage capacity, distributed energy storage access location, initial state of charge, and charge-discharge power limit, distributed power supply type, distributed power supply capacity, distributed power supply access location, node voltage historical measurement data, node injection power historical data, and node voltage reference value of each time period in m days Iteration step Δt, control step ΔT c , prediction step ΔT p , optimization duration T, and initialize control parameter k=1 and control time t=0;

[0008] Obtain node voltage data based on the power grid operation data and power grid operation parameters;

[0009] Obtain voltage deviation threshold value, and determine whether the voltage deviation of the node voltage data and the node voltage reference value exceeds the voltage deviation threshold value;

[0010] If yes, obtain node injection power measurement data based on the power grid operation data;

[0011] If no, set control time t=t+Δt.

[0012] Preferably, after the step of obtaining node injection power measurement data based on the power grid operation data, the method further includes:

[0013] Obtain the pseudo-Jacobian matrix of the distributed energy storage and the pseudo-Jacobian matrix of the distributed power supply based on the node injection power measurement data and the power grid operation data and power grid operation parameters;

[0014] Based on the pseudo-Jacobian matrix of the distributed energy storage, the pseudo-Jacobian matrix of the distributed power supply, and the power grid operation data and power grid operation parameters, a multi-level recursive algorithm is used to obtain a distributed energy storage pseudo-Jacobian estimation matrix predicted N steps from time t

[0015] constructing a first objective function based on a pseudo-Jacobi estimation matrix of the distributed energy storage a data-driven distributed energy storage active power adaptive prediction voltage control model is established based on the distributed energy storage state of charge constraint and the distributed energy storage active power output constraint, and a distributed energy storage active power output strategy at time t is obtained by solving.

[0016] Preferably, the step of constructing the first objective function is based on a pseudo-Jacobi estimation matrix of the distributed energy storage After the step of establishing a data-driven distributed energy storage active power adaptive prediction voltage control model based on the distributed energy storage state of charge constraint and the distributed energy storage active power output constraint, the method further comprises:

[0017] constructing a second objective function based on a distributed energy storage reactive power constraint, a distributed energy storage converter capacity constraint, a distributed power source reactive power constraint, and a distributed power source converter capacity constraint, establishing a data-driven distributed power source and energy storage coordinated voltage and reactive power control model to obtain a reactive power output strategy of the distributed power source at time t and the distributed energy storage at time t;

[0018] The distributed energy storage and the distributed power source output according to the reactive power output strategy of the distributed power source at time t and the distributed energy storage at time t, and the active power output strategy of the distributed energy storage at time t.

[0019] Preferably, after the step of the distributed energy storage and the distributed power source outputting according to the reactive power output strategy of the distributed power source at time t and the distributed energy storage at time t, and the active power output strategy of the distributed energy storage at time t, the method further comprises:

[0020] Let the control time t = t + Δt.

[0021] Preferably, after the step of letting the control time t = t + Δt, the method further comprises:

[0022] Determine whether t ≥ kΔT c is true,

[0023] If yes, let the prediction step ΔT p = ΔT p - ΔT c , update the control parameter k = k + 1, and determine whether the control time t is greater than the optimization time T.

[0024] If no, determine whether the control time t is greater than the optimization time T.

[0025] Preferably, after the step of determining whether the time t is greater than the optimization time length T, the method further comprises:

[0026] If not, the step of obtaining the voltage deviation threshold value, determining whether the voltage deviation between the node voltage data and the node voltage reference value exceeds the voltage deviation threshold value is executed.

[0027] If yes, the process ends.

[0028] Preferably, the distributed energy storage pseudo-Jacobian estimation matrix is:

[0029]

[0030] The iterative calculation expression is as follows:

[0031]

[0032] respectively represent the pseudo-Jacobian matrix of the lth distributed energy storage at time t and time t-Δt;

[0033] ΔU[t]=U[t]-U[t-Δt] represents the difference between the node voltage measurement at time t and time t-Δt;

[0034] represents the difference between the reactive power output of the lth distributed energy storage at time t-Δt and time t-2Δt;

[0035] η ESS and μ ESS represent weight coefficients;

[0036] represents the pseudo-Jacobian prediction matrix of the lth distributed energy storage at time t+rΔT c , r=1,…,N, and the calculation method is as follows:

[0037]

[0038] represents the pseudo-Jacobian prediction matrix of the lth distributed energy storage at time t+rΔT c (r=1,…,N);

[0039] θ j [t] represents the prediction coefficient at time t, where j=1,…,m, m represents the number of historical measurement data required when constructing the estimation sequence, T d represents the time interval of historical data, and ​respectively represent the pseudo-Jacobian matrix of the lth distributed energy storage at the same time of the previous m days calculated by the historical measurement data of the active power distribution network c -T d time, t+rΔT c -2T d time and t+rΔT c -mT d time, the pseudo-Jacobian matrix of the lth distributed energy storage; θ[t] = (θ1[t], …, θ m [t]) T , θ[t] is iteratively solved as follows:

[0040]

[0041] In the formula, δ represents the weight coefficient.

[0042] Preferably, the first objective function takes the minimum voltage deviation of each node and the lowest charging and discharging cost of the distributed energy storage as the control target, and specifically uses the following formula:

[0043]

[0044] the node voltage reference value vector of the future N steps starting from the time t+Δt;

[0045] the node voltage estimation value vector of the future N steps starting from the time t+Δt, wherein N represents the prediction step number, ΔT p is the prediction step length, and ΔT c is the ratio of the prediction step length ΔT

[0046]

[0047] E[t] represents a unit column vector, U[t] represents the measurement value of the node voltage of the power distribution network at the time t, represents the pseudo-Jacobian estimation matrix of the lth distributed energy storage for N steps from the time t;

[0048] the charging and discharging power change amount of the lth distributed energy storage at the time t;

[0049]

[0050] and respectively represent the charging and discharging power change amount vector of the lth distributed energy storage at the time t, t+ΔT c , and t+(N-1)ΔT c .

[0051] It represents the sum of the absolute values ​​of active charging and discharging power within N steps of the l-th distributed energy storage prediction;

[0052] λ represents the weighting coefficient;

[0053] The state-of-charge constraints for distributed energy storage are expressed as follows:

[0054]

[0055] SOC0—Initial state of charge of distributed energy storage;

[0056] SOC max and SOC min These represent the upper and lower limits of the state of charge, respectively;

[0057] This represents the converter capacity of the l-th distributed energy storage unit;

[0058] E[t] represents a unit column vector;

[0059] This represents the active power output of the l-th distributed energy storage at time t;

[0060] ΔT c To control the step size, N represents the number of prediction steps;

[0061] and They represent t-Δt+ΔT respectively c t-Δt+2ΔT c and t-Δt+NΔT c The change in the charging and discharging power of the distributed energy storage at time l;

[0062] SOC 0,l Let SOC represent the initial state of charge (SOC) of the l-th distributed energy storage device. T,l This represents the state of charge value after the l-th distributed energy storage unit has completed one operating cycle.

[0063] The active power output constraint of distributed energy storage is expressed as follows:

[0064]

[0065] These represent the upper and lower limits of the active power output of the l-th distributed energy storage, respectively.

[0066] E[t] represents a unit column vector;

[0067] This represents the active power output of the l-th distributed energy storage at time t.

[0068] Preferably, a second objective function is constructed, and a data-driven distributed power and energy storage coordinated voltage and reactive power control model is established based on the reactive power constraints of distributed energy storage, the capacity constraints of distributed energy storage converters, the reactive power constraints of distributed power sources, and the capacity constraints of distributed power source converters. This model includes:

[0069] The control objectives are to minimize node voltage deviation and minimize reactive power output deviation of distributed generation, specifically using the following formula:

[0070]

[0071] U ref —Voltage reference value, Let represent the reactive power output of distributed power source n at time t and time t-Δt, respectively;

[0072] Let represent the reactive power output of the l-th distributed energy storage at time t and time t-Δt, respectively;

[0073] λ DG and λ ESS Indicates the weighting coefficient;

[0074] The estimated voltage values ​​at each node at time t+Δt are expressed as follows:

[0075]

[0076] N DG N ESS These represent the number of distributed power sources and distributed energy storage devices, respectively.

[0077] —Estimated voltage values ​​of each node within time t+Δt;

[0078] U[t] — Voltage measurements at each node at time t;

[0079] —The pseudo-Jacobi matrix of the nth distributed source at time t;

[0080] —The pseudo-Jacobi matrix of the l-th distributed energy storage at time t;

[0081] Let represent the reactive power output of the nth distributed power source at time t and time t-Δt, respectively;

[0082] Let represent the reactive power output of the l-th distributed energy storage at time t and time t-Δt, respectively;

[0083] as well as The following formula is used:

[0084]

[0085] Let represent the pseudo-Jacobi matrices of the l-th distributed energy storage at time t and time t-Δt, respectively;

[0086] Let represent the pseudo-Jacobi matrices of the nth distributed source at time t and time t-Δt, respectively;

[0087] ΔU[t]=U[t]-U[t-Δt], representing the difference between the voltage measurements at each node at time t and time t-Δt; This represents the difference in reactive power output between the nth distributed power source at time t-Δt and time t-2Δt.

[0088] This represents the difference in reactive power output between the l-th distributed energy storage at time t-Δt and time t-2Δt.

[0089] η DG μ DG η ESS and μ ESS Indicates the weighting coefficient;

[0090] The reactive power output constraints and the capacity constraints of distributed energy storage converters are expressed as follows:

[0091]

[0092] —The upper and lower limits of the reactive power output of the nth distributed power source;

[0093] E[t] represents a unit column vector;

[0094] This represents the active power output of the nth distributed power source at time t;

[0095] This represents the capacity of the nth distributed power converter;

[0096] This represents the reactive power output of the nth distributed power source at time t;

[0097] The reactive power output strategies for distributed power generation and distributed energy storage at time t are obtained using the following formulas:

[0098]

[0099] U ref —Voltage reference value;

[0100] —Estimated voltage values ​​of each node at time t;

[0101] Let represent the pseudo-Jacobi matrix of the b-th distributed energy storage at time t;

[0102] Let represent the pseudo-Jacobi matrix of the a-th distributed source at time t;

[0103] Let represent the reactive power output of the nth distributed power source at time t and time t-Δt, respectively;

[0104] This represents the reactive power output of the l-th distributed energy storage at time t and time t-Δt. This represents the difference in reactive power output between the b-th distributed energy storage at time t-Δt and time t-2Δt.

[0105] N represents the difference in reactive power output between the a-th distributed power source at time t-Δt and time t-2Δt. DG N ESS These represent the number of distributed power sources and distributed energy storage devices, respectively.

[0106] ρ ESS and λ ESS —Weighting coefficient.

[0107] On the other hand, the present invention provides a data-driven distributed power supply and energy storage voltage control device, comprising:

[0108] The first acquisition module is used to acquire grid operation data and grid operation parameters. This data includes distributed energy storage type, capacity, location, initial state of charge, and charge / discharge power limits; distributed power source type, capacity, location; historical node voltage measurements, historical node injected power data, and node voltage reference values ​​for each time period over m days. Iteration step size Δt, control step size ΔT c Prediction step size ΔT p Optimize the duration T, and initialize the control parameter k=1 and the initial control time t=0;

[0109] The second acquisition module acquires node voltage data based on the power grid operation data and power grid operation parameters.

[0110] Decision module: Used to acquire voltage deviation threshold, determine the node voltage data and the node voltage reference value. Whether the voltage deviation exceeds the voltage deviation threshold.

[0111] The embodiments of the present invention bring the following beneficial effects: The present invention provides a data-driven distributed power source and energy storage voltage control method and device, the method comprising: acquiring grid operation data and grid operation parameters, and initializing control parameters k=1 and initializing control time t=0; acquiring node voltage data based on the grid operation data and grid operation parameters; acquiring a voltage deviation threshold, and determining the node voltage data and the node voltage reference value. If the voltage deviation exceeds the voltage deviation threshold, then the node injection power measurement data is obtained based on the grid operation data; otherwise, the control time t = t + Δt is set. The method provided by this invention can alleviate the technical problem that existing voltage regulation methods are affected by their discrete and slow adjustment methods, and cannot effectively cope with voltage fluctuations caused by the output uncertainty of distributed power sources. By dynamically characterizing the input-output relationship of the distribution network, establishing a data model, and then establishing a coordinated control model for distributed power sources and energy storage, the voltage control of the distribution network can be effectively realized, improving the safety of grid operation.

[0112] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention are realized and obtained in accordance with the structures particularly pointed out in the description, claims and drawings.

[0113] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0114] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0115] Figure 1 A flowchart of a data-driven distributed power supply and energy storage voltage control method provided in an embodiment of the present invention;

[0116] Figure 2 An active distribution network example topology diagram for a data-driven distributed power source and energy storage voltage control method provided in an embodiment of the present invention;

[0117] Figure 3 This invention provides a data-driven distributed power source and energy storage voltage control method, which includes prediction curves of distributed power source output and load changes.

[0118] Figure 4 A comparison of voltage results before and after 24-hour 33-node control in Embodiment 2 of a data-driven distributed power supply and energy storage voltage control method provided by the present invention.

[0119] Figure 5 The graph shows the changes in active power charging and discharging power and reactive power output of 33-node distributed energy storage over 24 hours in Embodiment 2 of a data-driven distributed power and energy storage voltage control method provided by the present invention.

[0120] Figure 6 The reactive power output diagram of a 24-hour 29-node distributed power source in Embodiment 2 of a data-driven distributed power source and energy storage voltage control method provided in this invention.

[0121] Figure 7 A comparison of the maximum and minimum node voltages before and after 24-hour voltage control in Embodiment 2 of a data-driven distributed power supply and energy storage voltage control method provided by the present invention;

[0122] Figure 8 The diagram shows the 24-hour state of charge change of each distributed energy storage device in Embodiment 2 of a data-driven distributed power source and energy storage voltage control method provided by the present invention. Detailed Implementation

[0123] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0124] Currently, with the high penetration rate of renewable energy in distribution networks, the power quality and reliability of these networks are negatively impacted. Traditional regulation methods, such as on-load tap-changing transformers and capacitor banks, while mitigating voltage exceedances to some extent, are ineffective in addressing voltage fluctuations caused by the uncertainty of distributed generation output due to their discrete and slow regulation methods. Since distributed generation only generates rated active power for a portion of the time, the remaining capacity of the inverter can be used for continuous voltage regulation. Therefore, reactive power control of distributed generation is a way to achieve rapid voltage regulation. Simultaneously, distributed energy storage is widely integrated into distribution networks due to its ease of control and operational flexibility. The widespread application of distributed energy storage provides rapid response capacity for distribution networks, helps smooth voltage fluctuations, regulates the spatiotemporal distribution of energy, and effectively improves the controllability and flexibility of distribution network operation. However, without coordination mechanisms, conflicts may arise between the voltage regulation effects of distributed energy storage optimization strategies and distributed generation reactive power control strategies, leading to voltage degradation and negatively impacting the safe and stable operation of the distribution network. Therefore, coordinated control of distributed power sources and distributed energy storage is of great significance for ensuring the safe and economical operation of the power distribution network. Based on this, the data-driven distributed power source and energy storage voltage control method and device provided in this invention can alleviate the technical problem that the voltage regulation method in the prior art is affected by its discrete and slow adjustment mode and cannot effectively deal with the voltage fluctuation caused by the output uncertainty of distributed power sources. By dynamically characterizing the input-output relationship of the power distribution network, a data model is established, and then a coordinated control model of distributed power sources and energy storage is established, which can effectively realize the voltage control of the power distribution network and improve the safety of the power grid operation.

[0125] To facilitate understanding of this embodiment, a data-driven distributed power supply and energy storage voltage control method disclosed in this embodiment of the invention will be described in detail first.

[0126] Example 1:

[0127] Combination Figure 1 Embodiment 1 of the present invention provides a data-driven distributed power source and energy storage voltage control method, which specifically includes the following steps:

[0128] Acquire grid operation data and parameters, including distributed energy storage type, capacity, location, initial state of charge, and charge / discharge power limits; distributed power source type, capacity, location; historical node voltage measurements, historical node injected power data, and node voltage reference values ​​for each time period over m days. Iteration step size Δt, control step size ΔT c Prediction step size ΔR pOptimize the duration T, and initialize the control parameter k=1 and the initial control time t=0;

[0129] Node voltage data is obtained based on the power grid operation data and power grid operation parameters;

[0130] Obtain the voltage deviation threshold and determine the node voltage data and the node voltage reference value. Whether the voltage deviation exceeds the voltage deviation threshold;

[0131] If so, then obtain node injection power measurement data based on the power grid operation data;

[0132] If not, then let the control time t = t + Δt.

[0133] Preferably, after the step of obtaining node injection power measurement data based on the power grid operation data, the method further includes:

[0134] Based on the node-injected power measurement data, the grid operation data, and the grid operation parameters, the pseudo-Jacobi matrix of distributed energy storage and the pseudo-Jacobi matrix of distributed power generation are obtained.

[0135] Based on the pseudo-Jacobi matrix of the distributed energy storage, the pseudo-Jacobi matrix of the distributed power source, and the grid operation data and parameters, a multi-level hierarchical algorithm is used to obtain the distributed energy storage pseudo-Jacobi estimation matrix for N steps of prediction starting from time t.

[0136] Construct the first objective function based on the pseudo-Jacobi estimation matrix of distributed energy storage. By considering the state of charge constraints and active power output constraints of distributed energy storage, a data-driven adaptive predictive voltage control model for the active power of distributed energy storage is established, and the active power output strategy of distributed energy storage at time t is obtained by solving the model.

[0137] Preferably, the construction of the first objective function is based on the distributed energy storage pseudo-Jacobi estimation matrix. After establishing a data-driven adaptive predictive voltage control model for distributed energy storage active power, constraining the state of charge and active power output of distributed energy storage, and solving for the active power output strategy of distributed energy storage at time t, the method further includes:

[0138] A second objective function is constructed. Based on the reactive power constraints of distributed energy storage, the capacity constraints of distributed energy storage converters, the reactive power constraints of distributed power sources, and the capacity constraints of distributed power source converters, a data-driven voltage and reactive power control model for coordinated distributed power sources and energy storage is established to obtain the reactive power output strategies of distributed power sources and distributed energy storage at time t.

[0139] Distributed energy storage and distributed power sources output power according to the reactive power output strategy of distributed power source at time t, the reactive power output strategy of distributed energy storage at time t, and the active power output strategy of distributed energy storage at time t.

[0140] Preferably, after the step of the distributed energy storage and distributed power source outputting according to the reactive power output strategy of the distributed power source at time t, the reactive power output strategy of the distributed energy storage at time t, and the active power output strategy of the distributed energy storage at time t, the method further includes:

[0141] Let the control time t = t + Δt.

[0142] Preferably, after the step of setting the control time t = t + Δt, the method further includes:

[0143] Determine t≥kΔT c Whether it is valid,

[0144] If so, then let the prediction step size be ΔT. p =ΔT p -ΔT c Update the control parameter k = k + 1, and determine whether the control time t is greater than the optimization duration T.

[0145] If not, then determine whether the control time t is greater than the optimization duration T.

[0146] Preferably, after the step of determining whether the control time t is greater than the optimization duration T, the method further includes:

[0147] If not, then proceed with obtaining the voltage deviation threshold to determine the node voltage data and the node voltage reference value. The step of determining whether the voltage deviation exceeds the voltage deviation threshold;

[0148] If so, the process ends.

[0149] Preferably, the distributed energy storage pseudo-Jacobi estimation matrix for:

[0150]

[0151] The iterative calculation expression is as follows:

[0152]

[0153] Let represent the pseudo-Jacobi matrices of the l-th distributed energy storage at time t and time t-Δt, respectively;

[0154] ΔU[t]=U[t]-U[t-Δt], representing the difference between the voltage measurements at each node at time t and time t-Δt;

[0155] This represents the difference in reactive power output between the l-th distributed energy storage at time t-Δt and time t-2Δt.

[0156] η ESS and μ ESS Indicates the weighting coefficient;

[0157] Represents t+rΔT c The pseudo-Jacobi prediction matrix for the distributed energy storage at time l, r = 1, ..., N, is calculated as follows:

[0158]

[0159] Represents t+rΔT c The pseudo-Jacobi prediction matrix of the distributed energy storage at time l (r = 1, ..., N);

[0160] θ j [t] represents the prediction coefficient at time t, where j = 1, ..., m, m represents the number of days of historical measurement data required to construct the estimation sequence, and T d Indicates the time interval of historical data. and They represent t+rΔT calculated using historical measurement data of the active distribution network at the same time in the previous m days. c -T d Time, t+rΔT c -2T d Time and t+rΔT c -mT d The pseudo-Jacobi matrix of the l-th distributed energy storage at time l; θ[t] = (θ1[t],…,θ m [t]) T The iterative solution formula for θ[t] is as follows:

[0161]

[0162] In the formula, δ represents the weighting coefficient.

[0163] Preferably, the first objective function takes minimizing the voltage deviation of each node and minimizing the cost of distributed energy storage charging and discharging as the control objectives, specifically using the following formula:

[0164]

[0165] —A vector of node voltage reference values ​​for the next N steps starting from time t+Δt;

[0166] —A vector of estimated node voltages for the next N steps starting from time t+Δt, where N represents the number of prediction steps and ΔT is the prediction step size. p With control step size ΔT c The ratio is expressed as follows:

[0167]

[0168] E[t] represents a unit column vector, and U[t] represents the measured value of the distribution network node voltage at time t. This represents the l-th pseudo-Jacobi estimation matrix for distributed energy storage, predicted N steps from time t.

[0169] —The change in charging and discharging power of the l-th distributed energy storage at time t;

[0170]

[0171] and Representing time t and t+ΔT respectively c Time and t+(N-1)ΔT c The vector of the change in the charging and discharging power of the distributed energy storage at time l;

[0172] It represents the sum of the absolute values ​​of active charging and discharging power within N steps of the l-th distributed energy storage prediction;

[0173] λ represents the weighting coefficient;

[0174] The state-of-charge constraints for distributed energy storage are expressed as follows:

[0175]

[0176] SOC0—Initial state of charge of distributed energy storage;

[0177] SOC max and SOC min These represent the upper and lower limits of the state of charge, respectively;

[0178] This represents the converter capacity of the l-th distributed energy storage unit;

[0179] E[t] represents a unit column vector;

[0180] This represents the active power output of the l-th distributed energy storage at time t;

[0181] ΔT c To control the step size, N represents the number of prediction steps;

[0182] and They represent t-Δt+ΔT respectively c t-Δt+2ΔT c and t-Δt+NΔT c The change in the charging and discharging power of the distributed energy storage at time l;

[0183] SOC 0,l Let SOC represent the initial state of charge (SOC) of the l-th distributed energy storage device. T,l This represents the state of charge value after the l-th distributed energy storage unit has completed one operating cycle.

[0184] The active power output constraint of distributed energy storage is expressed as follows:

[0185]

[0186] These represent the upper and lower limits of the active power output of the l-th distributed energy storage, respectively.

[0187] E[t] represents a unit column vector;

[0188] This represents the active power output of the l-th distributed energy storage at time t.

[0189] Preferably, a second objective function is constructed, and a data-driven distributed power and energy storage coordinated voltage and reactive power control model is established based on the reactive power constraints of distributed energy storage, the capacity constraints of distributed energy storage converters, the reactive power constraints of distributed power sources, and the capacity constraints of distributed power source converters. This model includes:

[0190] The control objectives are to minimize node voltage deviation and minimize reactive power output deviation of distributed generation, specifically using the following formula:

[0191]

[0192] U ref —Voltage reference value, Let represent the reactive power output of distributed power source n at time t and time t-Δt, respectively;

[0193] Let represent the reactive power output of the l-th distributed energy storage at time t and time t-Δt, respectively;

[0194] λ DG and λ ESS Indicates the weighting coefficient;

[0195] The estimated voltage values ​​at each node at time t+Δt are expressed as follows:

[0196]

[0197] N DG N ESS These represent the number of distributed power sources and distributed energy storage devices, respectively.

[0198] —Estimated voltage values ​​of each node within time t+Δt;

[0199] U[t] — Voltage measurements at each node at time t;

[0200] —The pseudo-Jacobi matrix of the nth distributed source at time t;

[0201] —The pseudo-Jacobi matrix of the l-th distributed energy storage at time t;

[0202] Let represent the reactive power output of the nth distributed power source at time t and time t-Δt, respectively;

[0203] Let represent the reactive power output of the l-th distributed energy storage at time t and time t-Δt, respectively;

[0204] as well as The following formula is used:

[0205]

[0206] Let represent the pseudo-Jacobi matrices of the l-th distributed energy storage at time t and time t-Δt, respectively;

[0207] Let represent the pseudo-Jacobi matrices of the nth distributed source at time t and time t-Δt, respectively;

[0208] ΔU[t]=U[t]-U[t-Δt], representing the difference between the voltage measurements at each node at time t and time t-Δt; This represents the difference in reactive power output between the nth distributed power source at time t-Δt and time t-2Δt.

[0209] This represents the difference in reactive power output between the l-th distributed energy storage at time t-Δt and time t-2Δt.

[0210] η DG μ DG η ESS and μ ESS Indicates the weighting coefficient;

[0211] The reactive power output constraints and the capacity constraints of distributed energy storage converters are expressed as follows:

[0212]

[0213] —The upper and lower limits of the reactive power output of the nth distributed power source;

[0214] E[t] represents a unit column vector;

[0215] This represents the active power output of the nth distributed power source at time t;

[0216] This represents the capacity of the nth distributed power converter;

[0217] This represents the reactive power output of the nth distributed power source at time t;

[0218] The reactive power output strategies for distributed power generation and distributed energy storage at time t are obtained using the following formulas:

[0219]

[0220] U ref —Voltage reference value;

[0221] —Estimated voltage values ​​of each node at time t;

[0222] Let represent the pseudo-Jacobi matrix of the b-th distributed energy storage at time t;

[0223] Let represent the pseudo-Jacobi matrix of the a-th distributed source at time t;

[0224] Let represent the reactive power output of the nth distributed power source at time t and time t-Δt, respectively;

[0225] This represents the reactive power output of the l-th distributed energy storage at time t and time t-Δt. This represents the difference in reactive power output between the b-th distributed energy storage at time t-Δt and time t-2Δt.

[0226] N represents the difference in reactive power output between the a-th distributed power source at time t-Δt and time t-2Δt. DG N ESS These represent the number of distributed power sources and distributed energy storage devices, respectively.

[0227] ρ ESS and λESS —Weighting coefficient.

[0228] Example 2:

[0229] Combination Figures 2 to 8 To verify the effectiveness of this method, the following two control schemes are compared for the power distribution network:

[0230] Option 1: Without coordinating and controlling distributed power sources and distributed energy storage, the initial operating state of the active power distribution network is obtained;

[0231] Option 2: Adopt a data-driven distributed power source and energy storage coordinated voltage control method.

[0232] The computer hardware environment for performing the optimized calculations was an Intel(R) Core(TM) CPU i5-10210U with a clock speed of 1.6GHz and 16GB of memory; the software environment was Windows 10.

[0233] The active distribution network topology used in this embodiment is as follows: Figure 2 As shown in the figure. The predicted curves for distributed power generation output and load information are as follows. Figure 4 As shown. A comparison of voltage results before and after control of energy storage access node 33 over 24 hours is shown below. Figure 4 As shown, the changes in active power charging and discharging power and reactive power output of the 33-node distributed energy storage system over 24 hours are as follows: Figure 5 As shown. The reactive power output of the distributed power source in Area 2 during 24 hours is as follows: Figure 6 As shown in the figure. The comparison results of the maximum and minimum voltage values ​​of each node before and after 24-hour voltage control are as follows. Figure 7 As shown. From Figures 4 to 7 As can be seen, Scheme 2 can effectively regulate the voltage level of the distribution network in this embodiment. The changes in the state of charge of each distributed energy storage system over 24 hours are as follows: Figure 8 As shown. (Summary) Figures 4 to 8 It can be seen that the data-driven distributed power source and energy storage coordinated voltage control method can effectively solve the voltage optimization problem.

[0234] Example 3:

[0235] Embodiment 3 of the present invention provides a data-driven distributed power supply and energy storage voltage control device, comprising:

[0236] The first acquisition module is used to acquire grid operation data and grid operation parameters. This data includes distributed energy storage type, capacity, location, initial state of charge, and charge / discharge power limits; distributed power source type, capacity, location; historical node voltage measurements, historical node injected power data, and node voltage reference values ​​for each time period over m days. Iteration step size Δt, control step size ΔT c Prediction step size ΔT p Optimize the duration T, and initialize the control parameter k=1 and the initial control time t=0;

[0237] The second acquisition module acquires node voltage data based on the power grid operation data and power grid operation parameters.

[0238] Decision module: Used to acquire voltage deviation threshold, determine the node voltage data and the node voltage reference value. Whether the voltage deviation exceeds the voltage deviation threshold.

[0239] Unless otherwise specifically stated, the relative steps, numerical expressions, and values ​​of the components and steps described in these embodiments do not limit the scope of the invention.

[0240] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0241] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than those shown in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0242] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the system and apparatus described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0243] Furthermore, in the description of the embodiments of the present invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in the present invention based on the specific circumstances.

[0244] In the description of this invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0245] Finally, it should be noted that the above-described embodiments are merely specific implementations of the present invention, used to illustrate the technical solutions of the present invention, and not to limit it. The scope of protection of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments within the technical scope disclosed in the present invention, or make equivalent substitutions for some of the technical features; and these modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A data-driven distributed power source and energy storage voltage control method, characterized in that, Specifically, the steps include the following: Acquire grid operation data and parameters, including distributed energy storage type, capacity, location, initial state of charge, and charge / discharge power limits; distributed power source type, capacity, location; historical node voltage measurements, historical node injected power data, and node voltage reference values ​​for each time period over m days. Iteration step size Δt, control step size ΔT c Prediction step size ΔT p Optimize the duration T, and initialize the control parameter k=1 and the initial control time t=0; Node voltage data is obtained based on the power grid operation data and power grid operation parameters; Obtain the voltage deviation threshold and determine the node voltage data and the node voltage reference value. Whether the voltage deviation exceeds the voltage deviation threshold; If so, then obtain node injection power measurement data based on the power grid operation data; If not, then let the control time t = t + Δt; After the step of obtaining node injection power measurement data based on the power grid operation data, the method further includes: Based on the node-injected power measurement data, the grid operation data, and the grid operation parameters, the pseudo-Jacobi matrix of distributed energy storage and the pseudo-Jacobi matrix of distributed power generation are obtained. Based on the pseudo-Jacobi matrix of the distributed energy storage, the pseudo-Jacobi matrix of the distributed power source, and the grid operation data and parameters, a multi-level hierarchical algorithm is used to obtain the distributed energy storage pseudo-Jacobi estimation matrix for N steps of prediction starting from time t. Construct the first objective function based on the pseudo-Jacobi estimation matrix of distributed energy storage. Based on the state of charge constraints and active power output constraints of distributed energy storage, a data-driven adaptive predictive voltage control model for active power of distributed energy storage is established, and the active power output strategy of distributed energy storage at time t is obtained by solving the model. A second objective function is constructed. Based on the reactive power constraints of distributed energy storage, the capacity constraints of distributed energy storage converters, the reactive power constraints of distributed power sources, and the capacity constraints of distributed power source converters, a data-driven voltage and reactive power control model for coordinated distributed power sources and energy storage is established to obtain the reactive power output strategies of distributed power sources and distributed energy storage at time t. Distributed energy storage and distributed power sources output power according to the reactive power output strategy of distributed power source at time t, the reactive power output strategy of distributed energy storage at time t, and the active power output strategy of distributed energy storage at time t.

2. The method according to claim 1, characterized in that, After the step of distributed energy storage and distributed power generation outputting according to the reactive power output strategy of distributed power generation at time t, the reactive power output strategy of distributed energy storage at time t, and the active power output strategy of distributed energy storage at time t, the method further includes: Let the control time t = t + Δt.

3. The method according to claim 1 or 2, characterized in that, After the step of setting the control time t = t + Δt, the method further includes: Determine t≥kΔT c Whether it is valid, If so, then let the prediction step size be ΔT. p =ΔT p -ΔT c Update the control parameter k = k + 1, and determine whether the control time t is greater than the optimization duration T; If not, then determine whether the control time t is greater than the optimization duration T.

4. The method according to claim 3, characterized in that, After the step of determining whether the control time t is greater than the optimization duration T, the method further includes: If not, then proceed with obtaining the voltage deviation threshold to determine the node voltage data and the node voltage reference value. The step of determining whether the voltage deviation exceeds the voltage deviation threshold; If so, the process ends.

5. The method according to claim 1, characterized in that, The distributed energy storage pseudo-Jacobi estimation matrix for: The iterative calculation expression is as follows: Let represent the pseudo-Jacobi matrices of the l-th distributed energy storage at time t and time t-Δt, respectively; ΔU[t]=U[t]-U[t-Δt], representing the difference between the voltage measurements at each node at time t and time t-Δt; This represents the difference in reactive power output between the l-th distributed energy storage at time t-Δt and time t-2Δt. η ESS and μ ESS Indicates the weighting coefficient; Represents t+rΔT c The pseudo-Jacobi prediction matrix for the distributed energy storage at time l, r = 1, ..., N, is calculated as follows: θ j [t] represents the prediction coefficient at time t, where j = 1, ..., m, m represents the number of days of historical measurement data required to construct the estimation sequence, and T d Indicates the time interval of historical data. and They represent t+rΔT calculated using historical measurement data of the active distribution network at the same time in the previous m days. c -T d Time, t+rΔT c -2T d Time and t+rΔT c -mT d The pseudo-Jacobi matrix of the l-th distributed energy storage at time l; θ[t] = (θ1[t],…,θ m [t]) T The iterative solution formula for θ[t] is as follows: In the formula, δ represents the weighting coefficient.

6. The method according to claim 1, characterized in that, The first objective function takes minimizing the voltage deviation of each node and minimizing the charging and discharging cost of distributed energy storage as the control objectives, and specifically adopts the following formula: —A vector of node voltage reference values ​​for the next N steps starting from time t+Δt; —A vector of estimated node voltages for the next N steps starting from time t+Δt, where N represents the number of prediction steps and ΔT is the prediction step size. p With control step size ΔT c The ratio is expressed as follows: E[t] represents a unit column vector, and U[t] represents the measured value of the distribution network node voltage at time t. This represents the l-th pseudo-Jacobi estimation matrix for distributed energy storage, predicted N steps from time t. —The change in charging and discharging power of the l-th distributed energy storage at time t; and Representing time t and t+ΔT respectively c Time and t+(N-1)ΔT c The vector of the change in the charging and discharging power of the distributed energy storage at time l; It represents the sum of the absolute values ​​of active charging and discharging power within N steps of the l-th distributed energy storage prediction; λ represents the weighting coefficient; The state-of-charge constraints for distributed energy storage are expressed as follows: SOCIETY 0,l =SOC T,l SOC0—Initial state of charge of distributed energy storage; SOC max and SOC min These represent the upper and lower limits of the state of charge, respectively; This represents the converter capacity of the l-th distributed energy storage unit; E[t] represents a unit column vector; This represents the active power output of the l-th distributed energy storage at time t; ΔT c To control the step size, N represents the number of prediction steps; and They represent t-Δt+ΔT respectively c t-Δt+2ΔT c and t-Δt+NΔT c The change in the charging and discharging power of the distributed energy storage at time l; SOC 0,l Let SOC represent the initial state of charge (SOC) of the l-th distributed energy storage device. T,l This represents the state of charge value after the l-th distributed energy storage unit has completed one operating cycle; The active power output constraint of distributed energy storage is expressed as follows: These represent the upper and lower limits of the active power output of the l-th distributed energy storage, respectively. E[t] represents a unit column vector; This represents the active power output of the l-th distributed energy storage at time t.

7. The method according to claim 1, characterized in that... A second objective function is constructed. Based on the reactive power constraints of distributed energy storage, the capacity constraints of distributed energy storage converters, the reactive power constraints of distributed power sources, and the capacity constraints of distributed power source converters, a data-driven coordinated voltage and reactive power control model for distributed power sources and energy storage is established, including: The control objectives are to minimize node voltage deviation and minimize reactive power output deviation of distributed generation, specifically using the following formula: U ref —Voltage reference value, Let represent the reactive power output of distributed power source n at time t and time t-Δt, respectively; Let represent the reactive power output of the l-th distributed energy storage at time t and time t-Δt, respectively; λ DG and λ ESS Indicates the weighting coefficient; The estimated voltage values ​​at each node at time t+Δt are expressed as follows: N DG N ESS These represent the number of distributed power sources and distributed energy storage devices, respectively. —Estimated voltage values ​​of each node within time t+Δt; U[t] — Voltage measurements at each node at time t; —The pseudo-Jacobi matrix of the nth distributed source at time t; —The pseudo-Jacobi matrix of the l-th distributed energy storage at time t; Let represent the reactive power output of the nth distributed power source at time t and time t-Δt, respectively; Let represent the reactive power output of the l-th distributed energy storage at time t and time t-Δt, respectively; as well as The following formula is used: Let represent the pseudo-Jacobi matrices of the l-th distributed energy storage at time t and time t-Δt, respectively; Let represent the pseudo-Jacobi matrices of the nth distributed source at time t and time t-Δt, respectively; ΔU[t]=U[t]-U[t-Δt], representing the difference between the voltage measurements at each node at time t and time t-Δt; This represents the difference in reactive power output between the nth distributed power source at time t-Δt and time t-2Δt. This represents the difference in reactive power output between the l-th distributed energy storage at time t-Δt and time t-2Δt. η DG μ DG η ESS and μ ESS Indicates the weighting coefficient; The reactive power output constraints and the capacity constraints of distributed energy storage converters are expressed as follows: This represents the converter capacity of the l-th distributed energy storage unit; This represents the active power output of the l-th distributed energy storage at time t; —The upper and lower limits of the reactive power output of the nth distributed power source; E[t] represents a unit column vector; This represents the active power output of the nth distributed power source at time t; This represents the capacity of the nth distributed power converter; This represents the reactive power output of the nth distributed power source at time t; The reactive power output strategies for distributed power generation and distributed energy storage at time t are obtained using the following formulas: U ref —Voltage reference value; —Estimated voltage values ​​of each node at time t; Let represent the pseudo-Jacobi matrix of the b-th distributed energy storage at time t; Let represent the pseudo-Jacobi matrix of the a-th distributed source at time t; Let represent the reactive power output of the nth distributed power source at time t and time t-Δt, respectively; This represents the reactive power output of the l-th distributed energy storage at time t and time t-Δt. This represents the difference in reactive power output between the b-th distributed energy storage at time t-Δt and time t-2Δt. N represents the difference in reactive power output between the a-th distributed power source at time t-Δt and time t-2Δt. DG N ESS These represent the number of distributed power sources and distributed energy storage devices, respectively. ρ ESS and λ ESS —Weighting coefficient.

8. A data-driven distributed power supply and energy storage voltage control device, characterized in that, include: The first acquisition module is used to acquire grid operation data and grid operation parameters. This data includes distributed energy storage type, capacity, location, initial state of charge, and charge / discharge power limits; distributed power source type, capacity, location; historical node voltage measurements, historical node injected power data, and node voltage reference values ​​for each time period over m days. Iteration step size Δt, control step size ΔT c Prediction step size ΔT p Optimize the duration T, and initialize the control parameter k=1 and the initial control time t=0; The second acquisition module acquires node voltage data based on the power grid operation data and power grid operation parameters. Decision module: Used to acquire voltage deviation threshold, determine the node voltage data and the node voltage reference value. Whether the voltage deviation exceeds the voltage deviation threshold; If so, then obtain node injection power measurement data based on the power grid operation data; If not, then let the control time t = t + Δt; After obtaining the node injection power measurement data based on the power grid operation data, the process also includes: Based on the node-injected power measurement data, the grid operation data, and the grid operation parameters, the pseudo-Jacobi matrix of distributed energy storage and the pseudo-Jacobi matrix of distributed power generation are obtained. Based on the pseudo-Jacobi matrix of the distributed energy storage, the pseudo-Jacobi matrix of the distributed power source, and the grid operation data and parameters, a multi-level hierarchical algorithm is used to obtain the distributed energy storage pseudo-Jacobi estimation matrix for N steps of prediction starting from time t. Construct the first objective function based on the pseudo-Jacobi estimation matrix of distributed energy storage. Based on the state of charge constraints and active power output constraints of distributed energy storage, a data-driven adaptive predictive voltage control model for active power of distributed energy storage is established, and the active power output strategy of distributed energy storage at time t is obtained by solving the model. A second objective function is constructed. Based on the reactive power constraints of distributed energy storage, the capacity constraints of distributed energy storage converters, the reactive power constraints of distributed power sources, and the capacity constraints of distributed power source converters, a data-driven voltage and reactive power control model for coordinated distributed power sources and energy storage is established to obtain the reactive power output strategies of distributed power sources and distributed energy storage at time t. Distributed energy storage and distributed power sources output power according to the reactive power output strategy of distributed power source at time t, the reactive power output strategy of distributed energy storage at time t, and the active power output strategy of distributed energy storage at time t.

Images