Measurement-driven new energy cluster distributed collaborative voltage control method, system and device and medium

Abstract

The invention discloses a measurement-driven new energy cluster distributed collaborative voltage control method, system and device and a medium, and belongs to the technical field of new energy cluster control. The method comprises the following steps: deploying a local controller in a plurality of distributed power supply clusters to collect voltage and power data; when the voltage deviation exceeds the limit, an autonomous and coordinated mapping matrix is constructed, a voltage control model is established, and a reactive power control strategy is optimized and solved; and if the local control is invalid, exchanging boundary information among the clusters and carrying out collaborative optimization to realize voltage regulation. The method has the advantages that the method adapts to new energy output uncertainty and power grid parameter unknown, local rapid adjustment is carried out firstly, cluster cooperative control is implemented when control cannot be carried out, local response efficiency and global coordination are considered, and the operation stability and the intelligent level of the power distribution network are improved.

Description

Technical Field

[0001] This invention relates to the field of new energy cluster control technology, specifically to a measurement-driven distributed collaborative voltage control method, system, device, and medium for new energy clusters. Background Technology

[0002] With the increasing integration of distributed generation into distribution networks, these networks are facing increasingly severe voltage fluctuation problems, which directly affect the power quality and safe operation of the distribution network. Distributed generation inverters can provide rapid reactive power support, offering an effective technical means to address voltage fluctuation issues.

[0003] However, most current reactive power control methods for distributed generation rely on precise line parameters, which are often difficult to obtain accurately in actual active distribution networks. This poses a challenge to control methods based on precise physical models in practical applications. Furthermore, the high proportion of distributed generation in the network makes the power flow distribution in active distribution networks more complex. Even if precise line parameters are available, it may be impossible to establish an accurate physical model to fully characterize the network's operating features. Even if a relatively accurate physical model can be established, the numerous complex components and frequent power flow changes in active distribution networks can make the model extremely complex, leading to difficulties in rapid solution and an inability to quickly respond to distributed generation fluctuations. Therefore, there is an urgent need to explore a more intelligent control method to achieve voltage adaptive control in the absence of precise physical parameters.

[0004] With the widespread deployment of advanced measurement devices in active power distribution networks, these networks can acquire a large amount of real-time measurement data. This data contains rich information on distribution network operation characteristics and user data, providing an important foundation for data-driven voltage control. Fully utilizing this measurement data to establish a data-driven voltage control model has become a highly promising control method, effectively addressing the problems associated with control methods based on precise physical models.

[0005] Distributed control methods rely solely on local measurement data and limited communication between clusters to formulate control strategies, characterized by small computational scale and low communication burden. This approach is particularly suitable for rapid reactive power control of distributed generation sources, enabling quick responses to voltage changes caused by distributed generation fluctuations, and providing a more flexible solution for voltage control in active distribution networks.

[0006] Based on the above background, this invention proposes a measurement-driven distributed collaborative voltage control method for new energy clusters. It aims to make full use of real-time measurement data to control the reactive power output of distributed power sources through intra-cluster autonomy and inter-cluster coordination, thereby achieving efficient and reliable voltage control and solving the problem of voltage fluctuation in distribution networks under high-proportion distributed power source access. Summary of the Invention

[0007] In view of the above-mentioned problems, the present invention is proposed.

[0008] Therefore, the technical problem solved by this invention is to provide a measurement-driven distributed collaborative voltage control method for new energy clusters when the precise physical parameters of the active power distribution network are unknown. This method formulates reactive power control strategies for distributed power sources through intra-cluster autonomy and inter-cluster coordination to quickly respond to voltage fluctuations caused by distributed power sources.

[0009] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a measurement-driven distributed collaborative voltage control method for new energy clusters, comprising,

[0010] Local controllers are deployed in multiple distributed power generation clusters to collect measurement data on node voltage and branch power within their respective clusters. When any cluster detects that the node voltage deviation exceeds a set threshold, its local controller constructs an autonomous mapping matrix for intra-cluster control and a coordination mapping matrix for inter-cluster collaboration based on the measurement data, and constructs a voltage control model accordingly. By optimizing and solving the voltage control model, the reactive power control strategy of the distributed power generation within the cluster is obtained. The reactive power control strategy is then distributed to the inverters of the distributed power generation within the cluster to adjust the reactive power output. If the adjusted node voltage still does not meet the predetermined requirements, the controllers of each cluster exchange boundary measurement information and update the coordination mapping matrix to construct an inter-cluster voltage control model that considers inter-cluster interaction. A distributed optimization method is used to solve the inter-cluster voltage control model to obtain the collaborative reactive power control strategy of the distributed power generation in each cluster, thereby achieving coordinated voltage regulation between clusters.

[0011] As a preferred embodiment of the measurement-driven distributed collaborative voltage control method for new energy clusters described in this invention, the construction of the autonomous mapping matrix includes controlling multiple distributed power sources within the cluster to sequentially adjust different reactive power output levels, collecting measurement data of the corresponding node voltage and branch power at each output level, and calculating mapping parameters reflecting the response of node voltage to reactive power regulation of the cluster based on the ratio between the node voltage change and reactive power output change at adjacent sampling times, so as to form an autonomous mapping matrix.

[0012] As a preferred embodiment of the measurement-driven distributed collaborative voltage control method for new energy clusters described in this invention, the construction of the coordination mapping matrix includes: collecting boundary node voltage and boundary branch power data of other clusters adjacent to this cluster to form extended voltage variables; associating the extended voltage variables with the reactive power output changes of this cluster; and establishing a coordination mapping matrix reflecting the voltage interaction relationship between clusters based on the corresponding response ratio.

[0013] As a preferred embodiment of the measurement-driven distributed collaborative voltage control method for new energy clusters described in this invention, the voltage control model is established based on an autonomous mapping matrix and a coordination mapping matrix. The optimization objective function is constructed with minimizing the deviation between the node voltage and the corresponding voltage reference value as the primary objective and minimizing the change amplitude of the reactive power output of the distributed power source as the secondary objective. The model is solved by an iterative optimization algorithm to obtain the reactive power control strategy within the cluster.

[0014] As a preferred embodiment of the measurement-driven distributed collaborative voltage control method for new energy clusters described in this invention, if the node voltage still does not meet the set safety range after the controller executes the reactive power control strategy to adjust the node voltage, the local controller and the controllers of adjacent clusters will exchange boundary information, including exchanging the voltage values ​​of boundary nodes and the active and reactive power information of boundary branches; and the coordination mapping matrix will be updated based on the latest boundary measurement data.

[0015] As a preferred embodiment of the measurement-driven distributed collaborative voltage control method for new energy clusters described in this invention, the method comprises: after updating the coordination mapping matrix, each local controller constructs a collaborative voltage control model that includes the voltage interaction effects between clusters. The model aims to minimize the combined voltage deviation and reactive power output change of each node, while considering constraints including power system power flow balance constraints, node voltage upper and lower limit constraints, distributed power source reactive power output upper and lower limit constraints, and cluster consensus constraints. A distributed optimization algorithm is used to solve the model to obtain the reactive power control strategy of each cluster for collaborative optimization.

[0016] As a preferred embodiment of the measurement-driven distributed collaborative voltage control method for new energy clusters described in this invention, in the iterative solution process of the collaborative voltage control model, each controller calculates the model residual based on the optimization results, including the original residual and the dual residual, and determines whether the consistency conditions between the clusters are met; if the consistency constraint threshold is met, it indicates that the clusters have reached a control consensus, and the iteration is terminated; if not, the boundary measurement information is exchanged and the consistency variables and dual variables are updated, and the optimization iteration is repeated until the control objective is achieved or the total control time ends.

[0017] Another objective of this invention is to provide a measurement-driven distributed collaborative voltage control system for new energy clusters.

[0018] To address the aforementioned technical problems, this invention provides the following technical solution: a measurement-driven distributed collaborative voltage control system for new energy clusters, comprising: multiple cluster-level local controllers, deployed in each distributed power generation cluster, used to collect measurement data of node voltage and branch power within their respective clusters, determine whether the node voltage exceeds a set threshold, and execute a voltage control strategy when trigger conditions are met; a data acquisition module, configured in each local controller, used to control multiple distributed power sources within the cluster to sequentially adjust reactive power output, and collect measurement data of node voltage and branch power at each output level; a mapping construction module, used to construct an autonomous mapping matrix and a coordinated mapping matrix based on measurement data at adjacent times, representing the response relationship of node voltage to reactive power regulation within their respective clusters, and the voltage interaction relationship between clusters; and an optimization solution module. The system is divided into four modules: a voltage control module and a consensus judgment module. The first module is used to construct a voltage control model based on a mapping matrix, aiming to minimize node voltage deviation and reactive power output variation. The second module is used to solve for the local reactive power control strategy through an optimization algorithm. The third module is used to execute inter-cluster cooperative voltage control when local control fails. This includes exchanging boundary node voltage and branch power measurement information with neighboring clusters, updating the coordination mapping matrix, constructing a voltage control model with inter-cluster interaction constraints, and solving for the cooperative reactive power control strategy using a distributed optimization method. The fourth module is used to distribute the reactive power control strategy obtained from local or cooperative control to the inverters of each distributed power source in the cluster to perform reactive power output adjustment. The fifth module is used to calculate the original residual and dual residual during the distributed cooperative optimization process, determine whether the control strategies of each cluster have converged to the consistency constraint, and control the termination of the iteration process.

[0019] The present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, characterized in that the processor executes the computer program to implement the steps of the measurement-driven distributed collaborative voltage control method for new energy clusters.

[0020] The present invention provides a computer-readable storage medium having a computer program stored thereon, characterized in that, when the computer program is executed by a processor, it implements the steps of the measurement-driven distributed collaborative voltage control method for new energy clusters.

[0021] The beneficial effects of this invention are: taking into account the unknown parameters of active power distribution network lines with a high proportion of distributed power sources and the uncertainty of the output of distributed power sources, a reactive power control strategy for distributed power sources is formulated through intra-group autonomy and inter-group coordination.

[0022] When distributed power sources cause voltage fluctuations, the voltage fluctuation cluster first uses local voltage measurement information to quickly build a measurement-driven autonomous voltage control model within the cluster in order to respond quickly to voltage fluctuations.

[0023] If intra-cluster autonomy cannot effectively control voltage, then distributed coordination is rapidly carried out among the clusters to establish a measurement-driven inter-cluster coordinated voltage control model. Through the cooperation between the clusters, the control effect is further optimized, which not only makes full use of local data, but also ensures the coordination and consistency of global control, effectively improving the operation and control level of the active distribution network. Attached Figure Description

[0024] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 A flowchart of a measurement-driven distributed collaborative voltage control method for new energy clusters provided in an embodiment of the present invention;

[0026] Figure 2 This is an improved IEEE 123-node distribution network structure diagram used in a measurement-driven distributed collaborative voltage control method for new energy clusters provided in an embodiment of the present invention.

[0027] Figure 3 A predicted curve of distributed power output of a measurement-driven distributed collaborative voltage control method for new energy clusters provided in an embodiment of the present invention;

[0028] Figure 4 A load information prediction curve for a measurement-driven distributed collaborative voltage control method for new energy clusters provided in an embodiment of the present invention;

[0029] Figure 5 The global voltage change diagram in scenario 1 (10:00-11:00) of the measurement-driven distributed collaborative voltage control method for new energy clusters provided in an embodiment of the present invention.

[0030] Figure 6 The global voltage change diagram in scenario 2 (10:00-11:00) of the measurement-driven distributed collaborative voltage control method for new energy clusters provided in an embodiment of the present invention.

[0031] Figure 7 The global voltage change diagram in scenario 3 (10:00-11:00) of the measurement-driven distributed collaborative voltage control method for new energy clusters provided in an embodiment of the present invention.

[0032] Figure 8A comparison of the maximum voltage values ​​in scenario one and scenario two from 10:00 to 11:00, which is a measurement-driven distributed collaborative voltage control method for new energy clusters provided in an embodiment of the present invention.

[0033] Figure 9 A comparison of minimum voltage values ​​in scenario one and scenario two from 10:00 to 11:00, for a measurement-driven distributed collaborative voltage control method for new energy clusters provided in an embodiment of the present invention.

[0034] Figure 10 The voltage coordination process diagram of cluster one and cluster three in scenario two of the measurement-driven distributed collaborative voltage control method for new energy clusters provided in an embodiment of the present invention (10:00-11:00).

[0035] Figure 11 The diagram shows the power coordination process of cluster one and cluster three in scenario two of the measurement-driven distributed coordinated voltage control method for new energy clusters provided in an embodiment of the present invention, from 10:00 to 11:00.

[0036] Figure 12 The diagram shows the change of the distributed power reactive power control strategy of cluster one in scenario two (10:00-11:00) of the measurement-driven distributed collaborative voltage control method for new energy clusters provided in an embodiment of the present invention.

[0037] Figure 13 The diagram shows the change of the distributed power reactive power control strategy of cluster three in scenario two (10:00-11:00) of the measurement-driven distributed collaborative voltage control method for new energy clusters provided in an embodiment of the present invention. Detailed Implementation

[0038] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0039] Example 1, referring to Figure 1 This is one embodiment of the present invention, which provides a measurement-driven distributed collaborative voltage control method for new energy clusters, comprising:

[0040] Local controllers are deployed in multiple distributed power clusters to collect measurement data of node voltage and branch power within the cluster.

[0041] When any cluster detects that the voltage deviation of a node within its own cluster exceeds a set threshold, its local controller constructs an autonomous mapping matrix for intra-cluster control and a coordination mapping matrix for inter-cluster collaboration based on the measurement data, and constructs a voltage control model accordingly.

[0042] By optimizing the voltage control model, the reactive power control strategy of the distributed power supply in this cluster is obtained.

[0043] The reactive power control strategy is distributed to the inverters of the distributed power sources within the cluster to adjust the reactive power output.

[0044] If the adjusted node voltage still does not meet the predetermined requirements, the controllers of each cluster exchange boundary measurement information and update the coordination mapping matrix to construct an inter-cluster voltage control model that considers inter-cluster interaction.

[0045] The distributed optimization method is used to solve the inter-cluster voltage control model to obtain the collaborative reactive power control strategy of each cluster's distributed power source, thereby realizing the collaborative voltage regulation between clusters.

[0046] The construction of the autonomous mapping matrix includes controlling multiple distributed power sources within the cluster to sequentially adjust different reactive power output levels, collecting measurement data of the corresponding node voltage and branch power at each output level, and calculating mapping parameters reflecting the relationship between node voltage and reactive power output at adjacent sampling times to form an autonomous mapping matrix.

[0047] In one optional embodiment of the present invention, the local controller controls multiple distributed power sources within the cluster to sequentially adjust their reactive power output. At each output level, voltage measurements are collected from the corresponding nodes. Data points are constructed using a finite number of adjustment levels, and the ratio between the voltage change and the reactive power output change is used as an approximate first-order sensitivity, forming an approximate autonomous mapping matrix. In this approach, each distributed power source performs only a finite number of perturbations, resulting in a smaller data volume and facilitating rapid estimation, but with relatively limited accuracy.

[0048] In a preferred embodiment of the present invention, in order to improve the accuracy of the autonomous mapping matrix, the controller adopts a multi-level disturbance strategy, that is, it controls multiple distributed power sources in the cluster to gradually adjust multiple discrete reactive power output levels within a preset range, for example, adjusting multiple times from -Qmax to +Qmax with equal interval steps.

[0049] At each reactive power output level, real-time measurement data of the corresponding node voltage and branch power are collected. For the sampling results between every two consecutive levels, the ratio between the voltage change and the reactive power output change is calculated, and finally a sensitivity mapping matrix of node voltage response to reactive power regulation of different distributed power sources is formed.

[0050] In this implementation, to reduce the impact of disturbances on actual operation, a rotating disturbance strategy can be adopted, that is, only a portion of the power supply is adjusted in each round to ensure system stability.

[0051] The beneficial effects of this preferred technical solution are as follows: by introducing a multi-level disturbance and rotation adjustment mechanism, the sensitivity calculation accuracy of the autonomous mapping matrix reflecting the relationship between voltage and reactive power output of each node can be significantly improved, and the adaptability of the model to nonlinear voltage-power response characteristics can be enhanced.

[0052] Meanwhile, with richer sampled data, the voltage control model constructed subsequently has stronger fitting ability and robustness, thereby improving the accuracy of reactive power control strategy and overall voltage regulation performance.

[0053] The construction of the coordination mapping matrix includes collecting boundary node voltage and boundary branch power data of other clusters adjacent to this cluster to form extended voltage variables;

[0054] The extended voltage variable is correlated with the reactive power output change of the cluster, and a coordination mapping matrix reflecting the voltage interaction relationship between the clusters is established based on the corresponding response ratio.

[0055] In one optional embodiment of the present invention, the local controller receives voltage measurements from boundary nodes and active and reactive power data from boundary branches from upstream or downstream clusters connected to it, and uses this as boundary information to construct extended voltage variables, which correspond one-to-one with the reactive power output changes of the local cluster. A simplified coordination mapping matrix is ​​then formed through ratio calculation. In this approach, the number of extended variables is limited, and the calculation is simple, but it does not fully cover all boundary dynamics.

[0056] In a preferred embodiment of the present invention, the construction of the coordination mapping matrix adopts a structured boundary modeling strategy. Specifically, the local controller first identifies all upstream and downstream clusters adjacent to its own cluster, collects the voltage values ​​of boundary nodes and the active and reactive power values ​​of boundary branches, and constructs these boundary information into a unified extended voltage variable set.

[0057] Subsequently, the controller compares the extended voltage set with the reactive power output change data of the current cluster, and calculates the response ratio between the changes in extended voltage and reactive power output over multiple consecutive time periods. To improve the stability of the calculation results, an exponentially weighted moving average method is preferably used to smooth the historical ratios, thereby forming a more accurate and disturbance-resistant coordination mapping matrix.

[0058] The beneficial effects of this preferred technical solution are: by introducing a dynamic modeling method with full boundary data, the coordination mapping matrix can accurately characterize the voltage response coupling relationship between clusters, thereby improving the structural expressive ability of the inter-cluster control model;

[0059] The exponential weighting strategy further enhances the ability to suppress fluctuation noise, improves the stability and practicality of the mapping matrix, and significantly improves the solution accuracy and convergence speed of subsequent inter-group collaborative optimization.

[0060] The voltage control model is established based on the autonomous mapping matrix and the coordination mapping matrix. It adopts the minimization of the deviation between the node voltage and the corresponding voltage reference value as the primary objective and the minimization of the change in reactive power output of distributed power sources as the secondary objective to construct an optimization objective function. The model is solved by an iterative optimization algorithm to obtain the reactive power control strategy within the cluster.

[0061] After the controller executes the reactive power control strategy to adjust the node voltage, if the node voltage still does not meet the set safety range, the local controller and the controllers of the adjacent clusters will exchange boundary information, including exchanging the voltage values ​​of the boundary nodes and the active and reactive power information of the boundary branches; and update the coordination mapping matrix based on the latest boundary measurement data.

[0062] In an optional embodiment of the present invention, the construction of the intra-group autonomous mapping matrix can be carried out using a perturbation injection-based identification method. Specifically, the local controller applies a perturbation with limited amplitude to the reactive power output of each distributed power source within a short time window, and collects the response curves of the node voltages during the perturbation. Then, using a least-squares fitting method, the sensitivity coefficients of each node voltage to the reactive power perturbation are extracted, thereby constructing an autonomous mapping matrix that reflects the impact of node voltage changes on reactive power output.

[0063] In a preferred embodiment of the present invention, a step-by-step reactive power disturbance injection method is adopted. That is, the controller controls the reactive power output of each distributed power source to change sequentially according to a preset order, typically selecting 3 to 5 discrete levels for step adjustment, and performing multiple voltage samples within the stabilization time of each level. By comparing the voltage change and reactive power output change between adjacent levels, the monotonic response trend of the node to reactive power output is calculated and normalized to improve the universality of the mapping relationship.

[0064] The advantages of this preferred technical solution are: it does not require a precise power grid parameter model, and can obtain a highly reliable node voltage response relationship solely through measurement data, thereby improving the model's practicality and robustness; at the same time, the step disturbance has good control feasibility, is suitable for engineering deployment, and will not have a significant impact on system stability.

[0065] After the updated coordination mapping matrix, each local controller constructs a collaborative voltage control model that includes the voltage interaction effects between groups. The model aims to minimize the combined voltage deviation and reactive power output change of each node, while considering constraints including power system power flow balance constraints, node voltage upper and lower limit constraints, distributed power source reactive power output upper and lower limit constraints, and inter-group consensus constraints.

[0066] A distributed optimization algorithm is used to solve the model and obtain the reactive power control strategy for collaborative optimization among the clusters.

[0067] During the iterative solution of the cooperative voltage control model, each controller calculates the model residual based on the optimization results, including the original residual and the dual residual, and determines whether the consistency conditions are met among the clusters.

[0068] If the consistency constraint threshold is met, it means that each cluster has reached a control consensus, and the iteration terminates; if not, the boundary measurement information is exchanged and the consistency variables and dual variables are updated, and the optimization iteration is repeated until the control objective is achieved or the total control time ends.

[0069] Example 2, an embodiment of the present invention, provides a measurement-driven distributed collaborative voltage control method for new energy clusters based on the previous embodiment, including:

[0070] Step 1) For a given set of multiple distributed renewable energy clusters, deploy a local controller for each cluster to formulate a reactive power control strategy for the distributed power sources within the cluster; input the parameter information of the local controller, specifically including: the access location, type and capacity of the distributed power sources within the cluster, the access location of the measurement devices within the cluster, the number of neighboring clusters, the number of boundary nodes and boundary branches, the sampling frequency f and the total control duration T; the local controller starts up as t=0;

[0071] It should be noted that, regarding step 1), the traditional method involves first establishing a physical model using specific parameters such as resistance, reactance, active and reactive power loads, and then solving it using a distributed algorithm. The physical model can be established using the DistFlow method. Considering the non-convex and nonlinear characteristics of the physical model, a second-order cone relaxation is further applied to the obtained model to reduce the solution complexity. The method in this embodiment does not rely on physical parameters, has a simple model, and is easy to solve, thus meeting the requirements of real-time control.

[0072] 2) Each cluster's local controller sequentially adjusts the reactive power output level of the distributed power source, samples the node voltage and branch power at each output level, and obtains cluster measurements; then, each cluster exchanges boundary information, specifically including: boundary node voltage and boundary branch power; based on the obtained cluster measurements and boundary information, the intra-cluster autonomous mapping matrix and inter-cluster coordination mapping matrix are initialized respectively; wherein:

[0073] In a preferred embodiment of the present invention, the specific definition of the cluster measurement and boundary information is as follows:

[0074]

[0075] In the formula, d is the cluster index; Let U be the set of voltages of node d in the cluster, where U i Let be the voltage at node i, where i is the index of the node. For the set of nodes in cluster d; Let P be the set of boundary node voltages of upstream neighboring clusters and boundary branch power of downstream neighboring clusters, representing the boundary information received by cluster d; where P l and Q l These are the active and reactive power of branch l, respectively, where l is the index of the branch; This represents the set of boundary nodes of upstream neighboring clusters of cluster d; This represents the set of boundary branches of downstream neighboring clusters of cluster d; The set of boundary branch power and boundary node voltage within cluster d represents the boundary information sent by cluster d. This represents the set of boundary nodes within cluster d. This represents the set of boundary branches within cluster d.

[0076] In a preferred embodiment of the present invention, for ease of model description, and The elements in the vector can be arranged into a column vector, as shown below:

[0077]

[0078] In the formula, y d It is a set A real vector composed of the elements in the array has a dimension of . This represents the node voltage of cluster d; It is a set A real vector composed of the elements in the array has a dimension of . This represents the reception boundary information of cluster d; It is a set A real vector composed of the elements in the array has a dimension of . This represents the transmission boundary information of cluster d.

[0079] In a preferred embodiment of the present invention, when defining y d , and Based on this, the initialization method for the intra-group autonomous mapping matrix is ​​as follows:

[0080] δ d [t]=Δ|y d [t]| / Δ|x d [t]|

[0081] In the formula, t represents the operation and control time of the new energy cluster, and δ d [t] is the intra-group autonomy mapping matrix of cluster d at time t; Δ|y d [t]|=|y d [t]-y d [t-Δt]| represents the absolute value of the voltage change at node d of the cluster from time t-Δt to time t; Δ|x d [t]|=|x d [t]-x d [t-Δt]| represents the absolute value of the change in reactive power control strategy of cluster d distributed power source from time t-Δt to time t.

[0082] In a preferred embodiment of the present invention, the initialization method of the inter-group coordination mapping matrix is ​​as follows:

[0083]

[0084] In the formula, For the extended voltage of cluster d; Let be the inter-group coordination mapping matrix of cluster d at time t; This represents the absolute value of the voltage change of cluster d from time t-Δt to time t.

[0085] In this preferred embodiment, the traditional method only defines voltage and does not consider the relationship between power between regions. The definition in this embodiment considers the relationship between voltage and power between upstream and downstream, which is more in line with the actual situation of the distribution network and accelerates the convergence between regions.

[0086] 3) Each cluster's measurement device samples node voltage and branch power; the local controller constructs voltage control trigger conditions based on the node voltage measurements; if a cluster meets the voltage control trigger conditions, proceed to the next step; otherwise, repeat step 3); where:

[0087] In a preferred embodiment of the present invention, the voltage control triggering condition is as follows:

[0088]

[0089] In the formula, The voltage reference value of cluster d at time t is represented, which is generally a column vector consisting of constants 1 and 0; ε represents the voltage control trigger threshold. The above formula indicates that if there exists a cluster d whose node voltage deviates from the voltage reference value by an absolute value greater than ε, then the voltage control triggering condition is met.

[0090] If the voltage control triggering condition is not met, the distributed generation adopts the reactive power control strategy of the previous moment and continues to sample the node voltage until the measurement meets the control triggering condition. The reactive power control strategy update method of the distributed generation when the voltage control triggering condition is not met is as follows:

[0091]

[0092] In the formula, x d [t+Δt] and x d [t] represents the reactive power control strategy of the cluster d distributed power source at time t+Δt and time t, respectively.

[0093] It should be noted that in step 3), the traditional method is to establish a physical model, while this method is a data-driven model. The traditional method model is not easy to establish and solve, and it is difficult to meet the needs of real-time control. This method only uses measurement data and has a linear relationship, so the model is easy to establish and solve, and meets the needs of real-time control.

[0094] 4) For clusters that meet the voltage control triggering conditions, their local controllers update the intra-cluster autonomous mapping matrix using the node voltage measurements obtained in step 3); where:

[0095] In a preferred embodiment of the present invention, the intra-group autonomous mapping matrix update method is as follows:

[0096]

[0097] In the formula, and Let represent the estimated intra-group autonomous mapping matrix of cluster d at time t and time t-Δt, respectively; Δy d [t] = y d [t]-y d [t-Δt] represents the change in node voltage of cluster d from time t-Δt to time t; Δx d [t-Δt]=x d [t-Δt]-x d [t-2Δt] represents the change in reactive power control strategy of cluster d from time t-2Δt to time t-Δt; μ represents the weight coefficient, used to penalize excessive changes in the autonomous mapping matrix within the cluster; η represents the step size factor.

[0098] It should be noted that the coefficients and thresholds are selected based on experience, with μ being greater than 0 and less than 1, and η being greater than 0 and less than 1.

[0099] It should be noted that in step 4), the traditional method does not consider changes in system state and uses a fixed time interval for control, which is time-triggered. The method in this embodiment sets voltage control trigger conditions, which is essentially event-triggered control. Control is only taken when the system state meets the trigger conditions, thus reducing the number of equipment actions while ensuring the safe operation of the distribution network.

[0100] 5) Based on the updated intra-cluster autonomous mapping matrix, the local controller establishes a measurement-driven intra-cluster autonomous voltage control model with the objective of minimizing node voltage deviation and overall reactive power output variation of distributed generation. The objective function is solved using the gradient descent method to obtain the reactive power control strategy for the distributed generation. This reactive power control strategy is then distributed to the inverters of the distributed generation, and each cluster measurement device acquires new node voltage and branch power measurements. Wherein:

[0101] In a preferred embodiment of the present invention, the objective function of the measurement-driven intra-group autonomous voltage control model is:

[0102]

[0103] in,

[0104] In the formula, F d (x d [t]) represents x d [t] is the objective function of the cluster d controlling the variables; Let be the estimated value of the node voltage of cluster d at time t+Δt; The voltage reference value of cluster d at time t+Δt; Δx d [t] = x d [t]-x d [t-Δt] represents the change in reactive power control strategy of cluster d from time t-Δt to time t; ω is the weighting coefficient, which is greater than 0.

[0105] The objective function is solved using the gradient descent method, yielding the following iterative solution formula for the reactive power control strategy of distributed generation:

[0106]

[0107] It should be noted that traditional methods establish physical models, while the method in this embodiment is different in three ways: first, it does not rely on physical parameters; second, the constraints in the model are linear, thus making the model simpler; and third, the strategy solution only involves simple single-number operations, resulting in a rapid solution.

[0108] 6) Based on the updated node voltage and branch power measurements, each cluster determines whether any cluster meets the voltage control triggering condition. If so, the local controllers of each cluster exchange boundary information, specifically including boundary node voltage and boundary branch power, and update the inter-cluster coordination mapping matrix; otherwise, return to step 3); where:

[0109] In a preferred embodiment of the present invention, the inter-group coordination mapping matrix update method is as follows:

[0110]

[0111] In the formula, and Let represent the estimated values ​​of the inter-group coordination mapping matrix of cluster d at time t and time t-Δt, respectively; This represents the change in the voltage of cluster d expansion from time t-Δt to time t.

[0112] It should be noted that in step 6), the traditional method incorporates boundary variables into the constraints of the physical model to establish a physical model for distributed voltage control. The method in this embodiment is characterized by three aspects: first, it does not rely on physical parameters; second, the constraints in the model are linear, thus making the model simpler; and third, this paper combines intra-group autonomy and inter-group coordination. Intra-group autonomy can quickly respond to source load fluctuations, while inter-group coordination can further improve the voltage control effect.

[0113] 7) Based on the updated inter-cluster coordination mapping matrix, each cluster's local controller aims to minimize the overall node voltage deviation and reactive power output variation of distributed generation. Considering system power flow constraints, node voltage safety constraints, distributed generation safety operation constraints, and cluster consensus constraints, a measurement-driven inter-cluster coordinated voltage control model is established. The model is solved in a distributed manner using the alternating direction multiplier method to obtain the reactive power control strategy for the distributed generation. The control strategy is then distributed to the inverters of the distributed generation, and each cluster's measurement devices acquire new node voltage and branch power measurements. Wherein:

[0114] In a preferred embodiment of the present invention, the objective function of the measurement-driven inter-group coordinated voltage control model is:

[0115]

[0116] In the formula, This represents the summation of the objective functions of each cluster.

[0117] The power flow constraints of the measurement-driven inter-group coordinated voltage control model are as follows:

[0118]

[0119] In the formula, Let be the estimated value of the expansion voltage of cluster d at time t+Δt, where and These are the estimated values ​​of the node voltage and the received boundary information of cluster d at time t+Δt, respectively.

[0120] The node voltage safety constraints are as follows:

[0121]

[0122] In the formula, and These are the lower and upper limits of the node voltage safety for cluster d at time t+Δt, respectively.

[0123] The constraints for the safe operation of distributed power sources are as follows:

[0124]

[0125] In the formula, x d [t] and These are the lower and upper limits of the reactive power output of the distributed power source of cluster d at time t, respectively.

[0126] The cluster consensus constraints are as follows:

[0127]

[0128] In the formula, z d [t] represents the consistency variable of cluster d at time t.

[0129] For the aforementioned measurement-driven inter-group coordinated voltage control model, each cluster's local controller employs an alternating direction multiplier method to solve the reactive power control strategy of the distributed power source. Taking the solution process of cluster d as an example, the explanation is as follows:

[0130] In a preferred embodiment of the present invention, the augmented Lagrangian function of the objective function of cluster d is first constructed as follows:

[0131]

[0132] In the formula, argmin represents the control policy x that returns the minimized augmented Lagrange function. d [t];λ d [t] represents the dual variable of cluster d at time t; ρ d Let be the update step size of the dual variable of cluster d.

[0133] Each cluster's local controller solves the augmented Lagrangian function in parallel, then distributes the reactive power control strategy of the distributed power source and acquires new cluster measurements; each cluster exchanges boundary information and then updates the consistency variables, as shown in the following equation:

[0134]

[0135] In the formula, z d [t+Δt] represents the consistency variable of cluster d at time t+Δt; This represents the reception boundary information of cluster d at time t+Δt.

[0136] Based on the updated consistency variables, the dual variables are updated as follows:

[0137]

[0138] In the formula, λ d [t+Δt] is the dual variable of cluster d at time t+Δt.

[0139] In step 7), traditional methods are either physically model-driven distributed or data-driven centralized; this method combines data-driven and distributed approaches, thus possessing the advantages of both.

[0140] 8) Based on the node voltage and branch power measurements obtained in step 7), calculate the original residual and dual residual of the measurement-driven inter-cluster coordinated voltage control model, and determine whether each cluster has reached a consensus; if a consensus is reached, proceed to the next step; otherwise, return to step 6); where:

[0141] In a preferred embodiment of the present invention, the calculation method for the original residual and dual residual of the measurement-driven inter-group coordinated voltage control model is as follows:

[0142]

[0143] υ d [t] = |z d [t]-z d [t-Δt]| / |z d [t]|

[0144] In the formula, τ d [t] and υ d [t] represents the original residual and dual residual of cluster d at time t, respectively; Let be the estimated value of the reception boundary information of cluster d at time t.

[0145] The condition for cluster consensus is that both the original residual and the dual residual converge to zero simultaneously, as shown in the following equation:

[0146]

[0147] In the formula, σ is the residual convergence threshold.

[0148] 9) Obtain the current control duration t = tk and determine the current control duration \(t\) k whether it is greater than the total control duration \(T\). If \(t\) k \(<T\), then go to step 3); otherwise, end.

[0149] It should be noted that in step 8), the combination of measurement-driven and distributed is a consistency judgment condition specifically proposed for the data-driven distributed framework and is an indispensable part of completing the control.

[0150] It should be noted that the traditional method is a distributed control method based on an accurate physical model. Although it can achieve distributed control, it depends on accurate physical parameters, which are often unavailable in actual distribution networks. In addition, the distributed control method based on an accurate physical model has a complex model and a slow convergence speed between regions, and cannot meet the requirements of real-time control. Traditional methods include the second-order cone relaxation technique.

[0151] Example 3, referring to Figures 2-13 This is an embodiment of the present invention, which provides a measurement-driven distributed collaborative voltage control method for new energy clusters. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through experiments.

[0152] For the embodiments of the present invention, 12 groups of distributed power sources are connected to the improved IEEE 123-node distribution network. Among them, the single installed capacity of the wind turbines at nodes 25, 35, and 46 is 600 kW, the single installed capacity of the wind turbines at nodes 15, 61, and 68 is 800 kW, the single installed capacity of the photovoltaic units at nodes 80, 88, 98, and 105 is 1000 kWp, and the single installed capacity of the photovoltaic units at nodes 114 and 121 is 1200 kWp. The system voltage level is 4.16 kV, and the active and reactive loads of the system are 3490 kW and 1920 kvar respectively. The measurement devices are deployed at all nodes of the distribution network, the sampling frequency \(f = 1\) Hz, and the total control duration \(T = 1\) h; the voltage reference value of the distribution network is set to 1.0 p.u. The measurement-driven distributed collaborative voltage control method for new energy clusters is used for control. Through the above steps, the reactive power control strategy of the distributed power sources can be obtained. To verify the effectiveness of this method, the following three control scenarios are used for comparison for the improved IEEE 123-node distribution network:

[0153] Scenario 1: Do not control the reactive power of the distributed power sources to obtain the initial operating state of the distribution network;

[0154] Scenario 2: Use the measurement-driven distributed collaborative voltage control method for new energy clusters;

[0155] Scenario 3: Use the distributed collaborative control method for new energy clusters based on an accurate physical model.

[0156] The computer hardware environment for performing the optimized calculations was an Intel(R) Core(TM) CPU i5-13500HX with a clock speed of 2.5GHz and 16GB of memory; the software environment was a Windows 11 operating system.

[0157] The improved IEEE 123-node distribution network structure used in this embodiment of the invention is as follows: Figure 2 As shown in the figure. The predicted output curve of the distributed power source is as follows. Figure 3 As shown. The load information prediction curve is as follows. Figure 4 As shown. The global voltage change in Scenario 1 from 10:00 to 11:00 is as follows. Figure 5 As shown in the figure. The global voltage change diagram in Scenario 2 from 10:00 to 11:00 is as follows. Figure 6 As shown. The global voltage change in Scenario 3 from 10:00 to 11:00 is as follows. Figure 7 As shown. Comparison of maximum voltage values ​​in Scene 1 and Scene 2 between 10:00-11:00. Figure 8 As shown. The minimum voltage values ​​in Scenario 1 and Scenario 2 between 10:00-11:00 are compared. Figure 9 As shown. The voltage coordination process between cluster 1 and cluster 3 in scenario 2 from 10:00 to 11:00 is as follows: Figure 10 As shown. The power coordination process between cluster 1 and cluster 3 in scenario 2 from 10:00 to 11:00 is as follows: Figure 11 As shown. The changes in the distributed power reactive power control strategy of cluster one in scenario two from 10:00 to 11:00 are as follows: Figure 12 As shown. The changes in the distributed power reactive power control strategy of cluster three in scenario two from 10:00 to 11:00 are as follows: Figure 13 As shown.

[0158] Depend on Figure 5 It can be seen that in Scenario 1, the initial operating state of the active distribution network has large voltage fluctuations and voltage over-limit occurs.

[0159] Depend on Figure 6 As can be seen, the measurement-driven distributed collaborative voltage control method for new energy clusters proposed in Scenario 2 effectively reduces voltage fluctuations and eliminates voltage exceedances. During the period of 10:00-11:00, the voltage of each node is effectively controlled within the range of 0.98-1.02. This is because the proposed method, based on measurement data feedback, can quickly detect voltage anomalies and initiate control rapidly. Furthermore, the proposed method only requires simple algebraic operations and solving a simple quadratic programming problem, resulting in a relatively small computational load, thus meeting the requirements of real-time control.

[0160] Depend on Figure 7As can be seen, the distributed collaborative control method for new energy clusters based on an accurate physical model in Scenario 3 cannot effectively control voltage and even worsens the initial operating state of the active distribution network. This is because the method based on an accurate physical model is computationally complex and has a long solution time. By the time the strategy is solved, the source load state has changed, causing the solved strategy to be inapplicable to the updated source load state. Therefore, it cannot effectively improve the voltage and may even worsen the initial operating state of the distribution network.

[0161] Depend on Figure 8 and Figure 9 It can be seen that the proposed method can effectively improve the maximum and minimum voltage values.

[0162] Depend on Figure 10 and Figure 11 It can be seen that the proposed method can achieve effective coordination among new energy clusters, thereby realizing voltage control.

[0163] By 12 and Figure 13 It can be seen that the proposed method can quickly adjust the reactive power output strategy of distributed power sources, thereby responding quickly to voltage fluctuations and rapidly improving voltage levels.

[0164] Example 4 is an embodiment of the present invention, which provides a measurement-driven distributed collaborative voltage control system for new energy clusters, including:

[0165] Multiple cluster-level local controllers are deployed in each distributed power cluster to collect measurement data of node voltage and branch power within the cluster, determine whether the node voltage exceeds the set threshold, and execute voltage control strategies when the trigger conditions are met.

[0166] The data acquisition module, configured in each local controller, is used to control multiple distributed power sources in the cluster to adjust reactive power output sequentially and to collect node voltage and branch power measurement data at each output level.

[0167] The mapping construction module is used to construct autonomous mapping matrices and coordinated mapping matrices based on measurement data at adjacent times, which respectively characterize the response relationship of node voltage to reactive power regulation within the cluster and the voltage interaction relationship between clusters.

[0168] The optimization solution module is used to construct a voltage control model based on the mapping matrix, with the goal of minimizing node voltage deviation and reactive power output variation, and obtain the local reactive power control strategy by solving through optimization algorithms.

[0169] The collaborative control module is used to perform inter-cluster collaborative voltage control when local control fails. This includes exchanging boundary node voltage and branch power measurement information with neighboring clusters, updating the coordination mapping matrix, constructing a voltage control model that includes inter-cluster interaction constraints, and using a distributed optimization method to solve the collaborative reactive power control strategy.

[0170] The execution and distribution module is used to distribute the reactive power control strategy obtained from local or collaborative control to the inverters of each distributed power source in the cluster to perform reactive power output adjustment.

[0171] The consensus judgment module is used to calculate the original residual and dual residual during the distributed collaborative optimization process, determine whether the control strategies among the clusters have converged to the consistency constraint, and control the termination of the iteration process.

[0172] This embodiment also provides an electronic device applicable to the measurement-driven distributed collaborative voltage control method for new energy clusters, comprising: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the measurement-driven distributed collaborative voltage control method for new energy clusters as proposed in the above embodiment.

[0173] This embodiment also provides a storage medium storing a computer program that, when executed by a processor, implements the measurement-driven distributed collaborative voltage control method for new energy clusters as proposed in the above embodiments.

[0174] The storage medium proposed in this embodiment and the measurement-driven distributed collaborative voltage control method for new energy clusters proposed in the above embodiments belong to the same inventive concept. Technical details not described in detail in this embodiment can be found in the above embodiments, and this embodiment has the same beneficial effects as the above embodiments.

[0175] Based on the above description of the implementation methods, those skilled in the art can clearly understand that the present invention can be implemented using software and necessary general-purpose hardware, and of course, it can also be implemented using hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk, or optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of the various embodiments of the present invention.

[0176] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A measurement-driven distributed collaborative voltage control method for new energy clusters, characterized in that: include, Local controllers are deployed in multiple distributed power clusters to collect measurement data of node voltage and branch power within the cluster. When any cluster detects that the voltage deviation of a node within its own cluster exceeds a set threshold, its local controller constructs an autonomous mapping matrix for intra-cluster control and a coordination mapping matrix for inter-cluster collaboration based on the measurement data, and constructs a voltage control model accordingly. By optimizing the voltage control model, the reactive power control strategy of the distributed power supply in this cluster is obtained. The reactive power control strategy is distributed to the inverters of the distributed power sources within the cluster to adjust the reactive power output. If the adjusted node voltage still does not meet the predetermined requirements, the controllers of each cluster exchange boundary measurement information and update the coordination mapping matrix to construct an inter-cluster voltage control model that considers inter-cluster interaction. The distributed optimization method is used to solve the inter-cluster voltage control model to obtain the collaborative reactive power control strategy of each cluster's distributed power source, thereby realizing the collaborative voltage regulation between clusters.

2. The measurement-driven distributed collaborative voltage control method for new energy clusters as described in claim 1, characterized in that: The construction of the autonomous mapping matrix includes controlling multiple distributed power sources within the cluster to sequentially adjust different reactive power output levels, collecting measurement data of the corresponding node voltage and branch power at each output level, and calculating mapping parameters reflecting the relationship between node voltage and reactive power output at adjacent sampling times to form an autonomous mapping matrix.

3. The measurement-driven distributed collaborative voltage control method for new energy clusters as described in claim 2, characterized in that: The construction of the coordination mapping matrix includes collecting boundary node voltage and boundary branch power data of other clusters adjacent to this cluster to form extended voltage variables; The extended voltage variable is correlated with the reactive power output change of the cluster, and a coordination mapping matrix reflecting the voltage interaction relationship between the clusters is established based on the corresponding response ratio.

4. The measurement-driven distributed collaborative voltage control method for new energy clusters as described in claim 3, characterized in that: The voltage control model is established based on the autonomous mapping matrix and the coordination mapping matrix. It adopts the minimization of the deviation between the node voltage and the corresponding voltage reference value as the primary objective and the minimization of the change in reactive power output of distributed power sources as the secondary objective to construct an optimization objective function. The model is solved by an iterative optimization algorithm to obtain the reactive power control strategy within the cluster.

5. The measurement-driven distributed collaborative voltage control method for new energy clusters as described in claim 4, characterized in that: After the controller executes the reactive power control strategy to adjust the node voltage, if the node voltage still does not meet the set safety range, the local controller and the controllers of the adjacent clusters will exchange boundary information, including exchanging the voltage values ​​of the boundary nodes and the active and reactive power information of the boundary branches; and update the coordination mapping matrix based on the latest boundary measurement data.

6. The measurement-driven distributed collaborative voltage control method for new energy clusters as described in claim 4, characterized in that: After the updated coordination mapping matrix, each local controller constructs a collaborative voltage control model that includes the voltage interaction effects between groups. The model aims to minimize the combined voltage deviation and reactive power output change of each node, while considering constraints including power system power flow balance constraints, node voltage upper and lower limit constraints, distributed power source reactive power output upper and lower limit constraints, and inter-group consensus constraints. A distributed optimization algorithm is used to solve the model and obtain the reactive power control strategy for collaborative optimization among the clusters.

7. The measurement-driven distributed collaborative voltage control method for new energy clusters as described in claim 4, characterized in that: During the iterative solution of the cooperative voltage control model, each controller calculates the model residual based on the optimization results, including the original residual and the dual residual, and determines whether the consistency conditions are met among the clusters. If the consistency constraint threshold is met, it means that each cluster has reached a control consensus, and the iteration terminates; if not, the boundary measurement information is exchanged and the consistency variables and dual variables are updated, and the optimization iteration is repeated until the control objective is achieved or the total control time ends.

8. A measurement-driven distributed collaborative voltage control system for new energy clusters, employing the measurement-driven distributed collaborative voltage control method for new energy clusters as described in any one of claims 1 to 7, characterized in that, include: Multiple cluster-level local controllers are deployed in each distributed power cluster to collect measurement data of node voltage and branch power within the cluster, determine whether the node voltage exceeds the set threshold, and execute voltage control strategies when the trigger conditions are met. The data acquisition module, configured in each local controller, is used to control multiple distributed power sources in the cluster to adjust reactive power output sequentially and to collect node voltage and branch power measurement data at each output level. The mapping construction module is used to construct autonomous mapping matrices and coordinated mapping matrices based on measurement data at adjacent times, which respectively characterize the response relationship of node voltage to reactive power regulation within the cluster and the voltage interaction relationship between clusters. The optimization solution module is used to construct a voltage control model based on the mapping matrix, with the goal of minimizing node voltage deviation and reactive power output variation, and obtain the local reactive power control strategy by solving through optimization algorithms. The collaborative control module is used to perform inter-cluster collaborative voltage control when local control fails. This includes exchanging boundary node voltage and branch power measurement information with neighboring clusters, updating the coordination mapping matrix, constructing a voltage control model that includes inter-cluster interaction constraints, and using a distributed optimization method to solve the collaborative reactive power control strategy. The execution and distribution module is used to distribute the reactive power control strategy obtained from local or collaborative control to the inverters of each distributed power source in the cluster to perform reactive power output adjustment. The consensus judgment module is used to calculate the original residual and dual residual during the distributed collaborative optimization process, determine whether the control strategies among the clusters have converged to the consistency constraint, and control the termination of the iteration process.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the measurement-driven distributed collaborative voltage control method for new energy clusters as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the measurement-driven distributed collaborative voltage control method for new energy clusters as described in any one of claims 1 to 7.

Images