Active distribution network double-layer voltage control method and system considering DG cluster access

Abstract

The present disclosure relates to the technical field of power grid voltage control, and provides a double-layer voltage control method and system for active distribution network considering DG cluster access, which is divided into two layers of network layer and distributed generation cluster layer control. In the network layer control of the upper layer, the reactive power output of the DGC in the active distribution network is taken as the control object, the different distributed generation clusters are coordinated, the stable control of the voltage of each node in the network layer is realized, and the requirement for communication performance and controller calculation capacity is reduced while the coordination control is performed. In the distributed generation cluster layer control of the lower layer, the reactive power output of each DG is taken as the control object, the reactive power distribution of each DG unit in the DGC is further optimized on the basis of the reactive power reference of the DGC obtained in the upper layer, and the deviation of the terminal voltage of each DG in the DGC is minimized.

Description

Technical Field

[0001] This disclosure relates to the technical field of power grid voltage control, specifically to a two-layer voltage control method and system for active distribution networks considering DG cluster access. Background Technology

[0002] The statements in this section are merely background information relating to this disclosure and do not necessarily constitute prior art.

[0003] The vigorous development of new energy power and the construction of new power systems are an inevitable trend. Distributed generation (DG) technology, with solar and wind power as the main energy sources, is developing rapidly, providing a possibility for reducing the output of traditional thermal power and solving severe energy and environmental problems. Especially for distribution networks, the scale of distributed photovoltaic power station access is constantly expanding, and its penetration rate is gradually increasing. The generated power can already meet the daily load demand of users and can even be fed back to the grid, which greatly alleviates the power supply pressure of traditional units. However, large-scale distributed power access also brings a series of power quality problems. In order to cope with the complex and ever-changing operating environment, the distribution network needs to have better regulation capabilities and gradually transform into a more flexible active distribution network. The output uncertainty of high-penetration DG in active distribution networks will cause frequent and significant fluctuations in distribution network voltage, and power backfeed will also cause the feeder voltage to rise, especially for low-voltage users, where the voltage fluctuations are more obvious. Therefore, optimizing the output of distributed energy within the active distribution network and studying voltage coordination control methods are particularly important.

[0004] According to the inventors, distributed generation (DG) connected to the grid based on power electronic interfaces can perform rapid and continuous power regulation and participate in voltage control in active distribution networks to control rapid voltage changes in the distribution network. There are three main voltage control strategies for active distribution networks with DG participation:

[0005] 1) Centralized Control Scheme. This scheme measures the overall information of the distribution network and uses various optimization control algorithms with different objective functions to coordinate the output of distributed generation (DG) to achieve voltage regulation. This method effectively coordinates the voltage regulation components in the active distribution network and achieves optimal voltage control. However, it requires monitoring of global information, thus placing high demands on the computing power and communication performance of the central controller.

[0006] 2) Distributed voltage control strategy. Also known as local control strategy, this is the simplest to implement. It uses local measurement data to optimize and control the distribution grid (DG) through droop control or power factor control. This control method does not require a central controller and has virtually no communication requirements, but it is difficult to implement various control components and cannot achieve optimal control of the overall distribution network voltage.

[0007] 3) Distributed voltage coordination control strategy. The optimal control problem is decomposed into multiple sub-problems and solved in a distributed controller, which reduces the computational burden on each controller and has a better control effect. However, the control effect of this strategy depends on the design of the distributed algorithm and the design of the coordination method between different components.

[0008] Distributed control is more suitable for voltage regulation in active distribution networks. The inventors found in their research that existing distributed voltage control methods only focus on voltage regulation at the medium-voltage distribution network level and do not consider voltage control of nodes within the low-voltage distributed generation unit cluster. That is, the low-voltage distributed generation cluster (DGC) is equivalent to a distributed power source issuing power commands, and then the issued power commands are evenly distributed to each distributed generation (DG) unit. The power allocation of each actual distributed generation unit within the DGC is not optimized, making it difficult to achieve voltage stability control that takes into account both the distribution network level and the distributed generation cluster level. Summary of the Invention

[0009] To address the aforementioned issues, this disclosure proposes a two-layer voltage control method and system for active distribution networks considering distributed generation (DG) cluster access. The system comprises two control layers: the distribution network layer and the distributed generation cluster layer. In the upper distribution network layer control, the reactive power output of the DGC in the active distribution network is used as the control object. By coordinating different distributed generation clusters, stable voltage control of each node in the distribution network layer is achieved, reducing the requirements for communication performance and controller computing power while maintaining coordinated control. In the lower distributed generation cluster layer control, the reactive power output of each DG is used as the control object. Based on the DGC reactive power reference obtained in the upper layer, the reactive power allocation of each DG unit is optimized, further realizing voltage control of the nodes within the DGC.

[0010] To achieve the above objectives, the present disclosure adopts the following technical solution:

[0011] One or more embodiments provide an active distribution network two-layer voltage control method considering DG cluster access, including two-layer control of the distribution network layer and the distributed generation cluster layer;

[0012] In the upper-level distribution network layer control, the reactive power output of the DGC in the active distribution network is taken as the control object. An objective function is constructed that includes the voltage stability requirements of each node in the distribution network layer and the reactive power margin requirements of the DGC. The optimal reactive power output of the DGC is obtained by solving the function, and distributed real-time control is performed on each DGC.

[0013] In the lower-level distributed generation cluster layer control, an objective function for the voltage stability requirements of the end nodes of the DG units within the DGC is constructed. Based on the reactive power output reference of the DGC obtained from the distribution network layer, the optimal reactive power output of each DG unit is obtained, and the DG units are then subjected to distributed control.

[0014] One or more embodiments provide an active distribution network two-layer voltage control system considering distributed generation cluster access, including a distribution network layer DGC control system and a DG control system. The DGC control system is set in the DGC controller to implement the control steps of the distribution network layer described above. The DG control system is set in the DG controller and configured to execute the control steps of the distributed generation cluster layer described above to realize the control of the distributed generation unit DG.

[0015] Compared with the prior art, the beneficial effects of this disclosure are as follows:

[0016] In this disclosure, the reactive power output of the distributed generation clusters (DGCs) in the active distribution network is used as the control object in the upper-level distribution network control. By coordinating different distributed generation clusters (DGCs), stable voltage control of each node in the distribution network layer is achieved. In the DGC layer control, the reactive power output of each DG is used as the control object. Based on the reactive power reference of the DGC obtained from the upper layer, the reactive power allocation of each DG unit is optimized to further achieve voltage control of the nodes within the DGC. This comprehensive approach considers the voltage coordination control of both the distribution network layer and the distributed generation cluster layer in the active distribution network. While optimizing the reactive power output of the DGC to ensure the minimum voltage deviation of each node in the distribution network layer, the reactive power allocation of each DG unit within the DGC is further optimized to minimize the voltage deviation of each DG terminal within the DGC.

[0017] The advantages of this disclosure, as well as its additional advantages, will be described in detail in the following specific embodiments. Attached Figure Description

[0018] The accompanying drawings, which form part of this disclosure, are used to provide a further understanding of this disclosure. The illustrative embodiments of this disclosure and their descriptions are used to explain this disclosure and do not constitute a limitation thereof.

[0019] Figure 1 This is a schematic diagram of the method flow of Embodiment 1 of this disclosure;

[0020] Figure 2 This is a typical active distribution network structure diagram provided by the method of Embodiment 1 of this disclosure;

[0021] Figure 3 This is a typical DGC structure diagram provided by the method of Embodiment 1 of this disclosure;

[0022] Figure 4 This is a voltage performance diagram of key nodes in the active distribution network before the application of the method in Embodiment 1 of this disclosure;

[0023] Figure 5 This is a voltage performance diagram of key nodes in an active distribution network after the application of the method in Embodiment 1 of this disclosure;

[0024] Figure 6This is a diagram showing the overall voltage performance of the active distribution network before and after the application of the method in Embodiment 1 of this disclosure;

[0025] Figure 7 This is a graph showing the overall voltage performance of DGC02 before and after applying the method of Embodiment 1 of this disclosure;

[0026] Figure 8 This is a graph showing the overall voltage performance of DG13 before and after applying the method of Embodiment 1 of this disclosure. Detailed Implementation

[0027] The present disclosure will be further described below with reference to the accompanying drawings and embodiments.

[0028] It should be noted that the following detailed descriptions are exemplary and intended to provide further illustration of this disclosure. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure pertains.

[0029] It should be noted that the terminology used herein is for descriptive purposes only and is not intended to limit the exemplary embodiments according to this disclosure. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof. It should be noted that, without conflict, the various embodiments and features within those embodiments can be combined with each other. The embodiments will now be described in detail with reference to the accompanying drawings.

[0030] Example 1

[0031] In one or more of the technical solutions disclosed in the embodiments, such as Figures 1-8 As shown, an active distribution network dual-layer voltage control method considering DG cluster access includes two layers of control: the distribution network layer and the distributed generation cluster layer.

[0032] In the upper-level distribution network layer control, namely distributed generation cluster (DGC) control: taking the reactive power output of the distributed generation cluster (DGC) in the active distribution network as the control object, constructing an objective function that includes the voltage stability requirements of each node in the distribution network layer and the reactive power margin requirements of the DGC, solving for the optimal reactive power output of the DGC, and performing decentralized real-time control on each DGC.

[0033] In the lower-level distributed generation cluster layer control, namely distributed generation unit control: construct the objective function of the voltage stability requirements of the end nodes of the internal nodes of the DGC, i.e., the DG unit, and solve for the optimal reactive power output of each DG unit based on the reactive power output reference obtained from the distribution network layer, and perform distributed control of the DG unit.

[0034] In this embodiment, the reactive power output of the distributed generation clusters (DGCs) in the active distribution network is used as the control object in the upper-level distribution network control. By coordinating different distributed generation clusters (DGCs), stable voltage control of each node in the distribution network layer is achieved. In the distributed generation cluster layer control, the reactive power output of each DG is used as the control object. Based on the reactive power reference of the DGC obtained from the upper layer, the reactive power allocation of each DG unit is optimized, further realizing voltage control of the nodes within the DGC. By comprehensively considering the voltage coordination control of the distribution network layer and the distributed generation cluster layer in the active distribution network, and while optimizing the reactive power output of the DGC to ensure the minimum voltage deviation of each node in the distribution network layer, the reactive power allocation of each DG unit within the DGC is further optimized, minimizing the voltage deviation of each DG terminal within the DGC.

[0035] In some possible implementations, after constructing the objective function in the distribution network layer control, i.e., the DGC control step, the following process is executed:

[0036] Step 11: Determine the reactive power constraints and network flow constraints imposed on DGC;

[0037] Step 12: Based on the objective function and constraints, determine the measurement values ​​required by the control method;

[0038] Specifically, the measured values ​​include the real-time voltage at the DGC access point and the real-time reactive power injection.

[0039] Step 13: Calculate the optimal reactive power compensation amount for DGC based on the obtained active distribution network measurement values ​​using a distributed gradient projection algorithm.

[0040] Step 14: Transmit the calculation results to the lower-level distributed generation cluster control for further calculation and allocation, and track the voltage fluctuation of the active distribution network in real time to achieve voltage stability of the active distribution network.

[0041] A further technical solution is to ensure the voltage stability of each node in the distribution network layer as follows: the voltage variation of each node in the active distribution network should be maintained within a certain range, and the closer it is to the voltage reference value, the better.

[0042] The reactive power margin requirement of the DGC in the distribution network layer control objective function is: the reactive power output of the distributed generation cluster in the active distribution network is close to the midpoint of the upper and lower limits of reactive power output, so as to ensure that it can cope with possible sudden disturbances.

[0043] Distributed generation clusters can be photovoltaic (PVC) clusters.

[0044] The following is a specific implementation example. In this example, the DGC control section of the distribution network layer includes the following steps:

[0045] 1.1) Construct an optimization objective function that includes voltage stability requirements and DGC reactive power margin requirements:

[0046] This disclosure can be applied to voltage control in active distribution networks. The main problem it addresses is voltage stability. This section regulates voltage by controlling the reactive power output of the distributed generation capacity (DGC). Therefore, a certain reactive power margin must be considered. The optimization objective function is to minimize the sum of the Euclidean norms of the differences between the voltage at each node of the active distribution network and the reactive power generated by the DGC and their respective standard values. The formula is as follows:

[0047]

[0048] Where q represents the reactive power input to the distribution network from each distributed generation cluster in the active distribution network; q mid Middle elements Defined as in q These represent the upper and lower limits of reactive power output that each distributed generation unit (DGC) can provide, determined by the specific processing conditions of the distributed generation unit; U and Ur are the actual and standard values ​​of the measured node voltage in the distribution network, respectively, where the standard value is represented by a per-unit value equal to 1; G U G q It is a weighting matrix, and G can be increased as the voltage deviation increases. U value.

[0049] The first term on the right side of the objective function for DGC control is to minimize the voltage deviation, and the second term is to minimize the reactive power deviation between the reactive power output of DGC and the intermediate value of reactive power output, that is, to ensure that DGC has a certain reactive power margin.

[0050] 1.2) Determine the reactive power output of the DGC and the operating constraints it should be subject to:

[0051] The operational constraints, namely the constraints of the objective function mentioned in 1) above, mainly include equality constraints considering the relationship between voltage and power flow in the distribution network, and inequality constraints considering the reactive power output limit of the DGC. Specifically, the constraints include that the reactive power compensation that the DGC can supply to the active distribution network should be within its upper and lower limits, which are determined by the upper and lower limits of the reactive power output of each DG unit within the DGC, and the power flow constraints representing the relationship between reactive power, active power, and voltage after the active distribution network is linearized.

[0052] The inequality constraint considering the reactive power output limit of the DGC is that the reactive power output lies between the upper and lower limits of the DGC reactive power output. Here, the reactive power constraint of DGC is obtained based on the upper and lower limits of reactive power output of each DG unit transmitted by the DG controller.

[0053] Considering the equality constraints between voltage and power flow in distribution networks, and due to the inherent characteristics of distribution networks, power losses on lines can be ignored during normal operation. The commonly used linear power flow model of distribution networks can be transformed into a matrix model from which voltage equality constraints are obtained:

[0054] -AP=-p (2)

[0055] -AQ=-q-BU j (3)

[0056]

[0057] in, This is the correlation matrix, representing the direct correlation between two nodes in the network. A0 corresponds to the network connection point (POC) in the first column, and A is the correlation matrix for the remaining nodes, which is a square matrix. The specific values ​​of the elements are determined using the same method as in existing methods. P, Q, p, and q correspond to the active power, reactive power transmitted between adjacent nodes in the distribution network, and the active power and reactive power injected into the nodes, respectively. U, R, and X represent the matrix form of node voltages and the various quantities of resistance and reactance between nodes. U0 represents the voltage at the POC, BU... j This represents the reactive power flow from the DGC to ground at the DGC grid connection point, where the order of the elements needs to correspond to the incidence matrix to be equivalent to commonly used distribution network power flow models. Thus, substituting the first two equations into the third equation yields the equality constraints:

[0058]

[0059] Among them, 2(A) Τ ) -1 XA -1 Defined as M, which is a symmetric positive definite matrix. Voltage performance without external reactive power injection:

[0060]

[0061] 1.3) Based on the objective function and constraints, determine the power distribution network measurement values ​​required to achieve the objective function;

[0062] The optimal reactive power output of the DGC is obtained by optimizing the calculation results using the gradient distributed solution algorithm based on the measured power distribution network data. The reactive power output for a future period of time is then optimized and the global variables are updated and passed to the DG controller.

[0063] For the DGC control section, the voltage stability requirement is that the voltage fluctuations at the distribution network nodes must be within a certain range, and the closer to the reference value, the better. The reactive power margin requirement is that the reactive power injected into the distribution network by the DGC must be close to the midpoint of its upper and lower reactive power output limits, so that the distribution network has sufficient coping ability when the network is disturbed.

[0064] The control algorithm described in the DGC control section performs distributed real-time control of each DGC, with no communication between adjacent DGC units. The control role of the DGC is considered at the distribution network layer, utilizing a fully distributed control method. Measurement data and optimal solution calculations are performed locally on the distributed cluster controller, obtaining a DGC reactive power output reference that maximizes voltage stability at each node in the distribution network layer. This approach places low demands on the controller's computational and communication performance, and offers fast solution speed.

[0065] Specifically, the process of achieving fully decentralized control includes:

[0066] Based on the overall optimal objective of the active distribution network, the optimal solution is sought in the direction of the negative gradient.

[0067] By scaling the gradient values ​​and setting the weight factor to a constant, the solution process is decoupled to be related only to local information measurements, thereby solving for the optimal solution of DG reactive power output.

[0068] In some embodiments, gradient scaling improves the algorithm's convergence rate by introducing a scaling matrix, and the rules for selecting weight factors are designed to decouple the optimal solution from the variables in the solution formula to a form independent of non-local information. This results in a completely decentralized solution method, where all processes can be completed by the DGC's internal control and measurement units.

[0069] In a specific embodiment, the control algorithm described in the DGC control section involves the DGC measuring the nodes directly connected to the DGC, and then calculating the reactive power reference that can affect the entire network and achieve overall voltage stability. The specific process includes:

[0070] Starting from the overall optimal objective, seek the optimal solution in the direction of the negative gradient;

[0071] By decoupling the solution of the iterative equations, the DGC controller can perform local calculations and control, thereby achieving overall network voltage stability.

[0072] Methods for solving iterative equations to handle decoupling mainly include:

[0073] A positive definite scaling matrix H is introduced to improve the convergence speed of the algorithm, and the value of the positive definite matrix is ​​set to achieve distributed decoupling.

[0074] Set the values ​​of the weight matrix to achieve distributed decoupling;

[0075] Set the value of the step size parameter to improve calculation speed.

[0076] The gradient dispersion algorithm is used to optimize the solution method, specifically according to the negative gradient direction. Iterate and update the solution, then map the updated solution to the constraint set. In the above solution, a positive definite scaling matrix H is added as the gradient coefficient, and a diagonal approximation of the Hessian matrix is ​​used. The step size parameter is set to a constant. The specific solution process is as follows:

[0077] First, regarding the objective function and constraints mentioned above, one solution method used here is the gradient method, that is, along the negative gradient direction. Iterate and update the solution, then map the updated solution to the constraint set. The iterative update formula for the gradient method is as follows:

[0078]

[0079] Where k represents the iteration step, α(k)∈(0,1] is the step size, γ(k>0) is a positive scalar, and [z] is the step size. + The method of mapping to a constraint set, for any vector z∈R N [z] + The value can be determined using the following rules:

[0080]

[0081] In this case, q(k+1) can remain within the constraints as long as The gradient of the objective function.

[0082] To achieve a fast discretization of the reactive power reference at the DGC output, the above solution is processed as follows:

[0083] Since updating and calculating γ and α in each iteration increases computation time, to improve computation speed, we keep α(k) = α and γ(k) = γ constant, that is:

[0084]

[0085] Since the original gradient method has a slow convergence speed, a scaling matrix H is introduced to improve the iterative convergence speed. H is a positive definite matrix, and the iterative equation obtained after scaling is as follows:

[0086]

[0087] To achieve distributed real-time voltage control, it is necessary to decouple the terms related to global information in the iterative equation into local information representations. For the above equation, this means decoupling the terms related to global information from the iterative equation. After processing, the global gradient information can be obtained from the local voltage U of each node. i (k) Voltage standard value U r And reactive power information q i (k) is obtained.

[0088] The gradient is obtained by solving the aforementioned objective function equation and constraint equations, when the weight matrix G is set. U =M -1 Gq = diag(g q1 ,K,g qN ), where g qN Let the positive elements be on the diagonal, and then... Substituting the values, we obtain the decoupled gradient as follows:

[0089]

[0090] To improve the convergence speed while satisfying the need for decoupling of the solution method, a diagonal approximation of the Hessian matrix is ​​used for H, namely:

[0091]

[0092] in, Here, "f0" indicates that the matrix is ​​positive definite.

[0093] After the above processing, the iterative solution formula is transformed into:

[0094]

[0095] In this way, the reactive power reference required for each DGC to achieve the optimal global voltage effect can be obtained locally at the DGC based solely on the information of its connected nodes.

[0096] In one possible implementation, after constructing the objective function in the lower-level distributed generation cluster layer control step, the following process is performed:

[0097] Step 21: Determine the reactive power output constraints and network power flow constraints imposed on each distributed generation unit.

[0098] Step 22: Based on the objective function and constraints, determine the required DG network data measurement values ​​for the solution;

[0099] Specifically, the measured values ​​include the real-time voltage at the DG access point and the real-time reactive power output of the DG unit.

[0100] Step 23: Solve the objective function using the alternating direction multiplier method (ADMM) based on the measurement data to obtain the optimal reactive power output of each DG;

[0101] Step 24: Send the optimization calculation results to the DG unit to achieve voltage stability control that takes into account both the distribution network layer and the distributed generation cluster layer.

[0102] Specifically, the process of implementing distributed control includes:

[0103] Step 221: Construct the corresponding Lagrangian function based on the objective function of the distributed generation cluster layer control;

[0104] Step 222: The DGC controller updates the global variables under the reactive power reference constraints obtained from the upper distribution network layer control calculation and passes them to the DG controller;

[0105] Step 223: The DG controller updates the local variables based on the updated global variables and the reactive power output inequality constraints, then updates the dual variables based on the updated values, and passes the updated variables to the DGC controller.

[0106] Step 224: Through the interaction between the DG controller and the DGC controller, the optimal reactive power output of the DG is obtained, thereby optimizing the voltage of each node within the DGC.

[0107] The following is a specific implementation example, focusing on the distributed generation cluster (DG) control section:

[0108] 2.1) Construct an optimization objective function that includes the voltage stability requirements of each DG end node in the DGC layer:

[0109] In this embodiment, while maintaining the stability of the distribution network layer voltage, the voltage of the nodes within the DGC is further controlled. This part regulates the DGC layer voltage by controlling the specific reactive power output of each DG unit within the DGC. Therefore, the optimization objective function is: to minimize the sum of the squared differences between the voltage of each DG terminal node and its corresponding standard value, as shown in the following formula:

[0110]

[0111] in, This represents the number of DG cells within DGC i. The objective function is to minimize the deviation of the DG terminal node voltage from the reference value, thereby keeping the node voltages within the DGC stable.

[0112] 2.2) Determine the reactive power output of the DG and the operating constraints it should be subject to:

[0113] The constraints on the objective function mentioned in Formula 12) mainly consist of three parts: equality constraints reflecting the relationship between the DGC layer voltage and the DG reactive power output change, inequality constraints considering the DG reactive power output limit, and equality constraints on the total DG reactive power output calculated by the upper-level DGC.

[0114] Specifically, the constraints include: the reactive power output of each DG should be between the upper and lower limits of its capacity, and the sum of the reactive power output of all DGs should meet the reactive power requirements of the DGC, as well as the voltage constraint on the relationship between the reactive power of the DGC network and the voltage, expressed in terms of voltage sensitivity.

[0115] Considering the inequality constraint on the reactive power output limit of the DG, it can be calculated based on the rated power of the DG unit and the measured real-time active power output, that is:

[0116]

[0117] Among them, the reactive power inequality constraints of each DG are output to the DGC controller to calculate the upper and lower limits of the reactive power output of the DGC.

[0118] The equation constraint reflecting the relationship between the DGC layer voltage and the reactive power output change of the DG, namely,

[0119]

[0120] in, The measured terminal voltage of each DG node in the DGC layer; ΔQ DG and These represent the vector of reactive power output change in the DG network and the sensitivity of the DGC network voltage to changes in reactive power output in the DG network, respectively. For reference voltage, 1.0 pu can be used.

[0121] Since voltage regulation of the active distribution network must first ensure voltage stability at each node of the distribution network layer, the sum of reactive power output of all DG units within the lower-level DGC should strictly track the reactive power output reference calculated by the upper-level DGC to minimize the voltage deviation of the distribution network layer. Therefore, the equation constraint is:

[0122]

[0123] in, Optimized reactive power output for each DG node in the DGC layer; To consider the sensitivity of network loss to changes in DG reactive power output when considering the reactive power loss of the DGC layer network; The reactive power reference obtained for the DGC controller is the value obtained from equation (11) above.

[0124] 2.3) Based on the objective function and constraints, the optimization problem is decomposed using the ADMM algorithm;

[0125] Specifically, the solution method is explained as follows:

[0126] Introducing a vector z composed of global variables zj, the constraint shown in equation (14) can be written as ΔQ.DG If -z = 0, then the augmented Lagrangian function can be expressed as:

[0127]

[0128] Where y represents the dual variable y j The vector formed by ρ; ρ is the penalty term of ADMM.

[0129] If feasible, the entire optimization problem can be solved according to the following steps:

[0130] The DGC controller solves an optimization problem with global constraints as follows:

[0131]

[0132] st(17)

[0133] The DG controller solves the locally constrained optimization problem based on the updated global variables and updates the local variable ΔQ. DGj and dual variable y j ;

[0134]

[0135]

[0136]

[0137] In this way, the overall optimization problem is decomposed into multiple sub-problems, which are solved by the DGC controller and the DG controller respectively, reducing the computational burden on each controller. After the iteration converges, the DGC controller sends the optimization results to the DG unit to control its reactive power output, so as to achieve voltage coordination control that takes into account both the distribution network layer and the internal voltage of the DGC unit.

[0138] To enable those skilled in the art to better understand the technical solutions of this application, the technical solutions of this embodiment will be described in detail below with reference to specific simulation examples and comparative examples.

[0139] The operation flow of the coordinated voltage control method applied to active distribution networks mentioned in this embodiment is as follows: Figure 1 As shown, this mainly illustrates the DGC control section of the distribution network layer, the DG control section of the DGC layer, and the corresponding communication paths. The specific control methods for each layer have been detailed previously. Figure 2 The typical power distribution network mentioned above is simulated in SIMULINK. The specific topology of the distributed generation cluster 2 connected to node 07 is as follows: Figure 3 As shown, it contains 20 distributed generation units. Figure 4 , Figure 5What is shown is the voltage information of key network nodes (nodes 5 and 45 connected to distributed generation cluster 1 and distributed generation cluster 8, and node 9 at the end of the feeder) before and after control under normal operating conditions, that is, when the active power injected into the distribution network by DGC changes. After control, the voltage fluctuation is significantly reduced. Figure 6 This demonstrates the improvement in overall distribution network voltage before and after the application of this disclosure. After applying the method of this embodiment, the voltage deviation is reduced. Figure 7 , Figure 8 This shows the overall voltage deviation of each distributed generation unit terminal node in the distributed generation cluster layer and the voltage status of distributed generation unit 13 terminal node in distributed generation cluster 2. It can be seen that the voltage stability is significantly improved after applying the method of this embodiment.

[0140] Example 2

[0141] Based on Embodiment 1, this embodiment provides an active distribution network dual-layer voltage control system considering DG cluster access, including a DGC control system at the distribution network layer and a DG control system at the DGC layer. The DGC control system is set in the DGC controller to implement the control steps of the distribution network layer described in Embodiment 1; the DG control system is set in the DG controller and configured to execute the control steps of the distributed generation cluster layer described in Embodiment 1 to realize the control of the distributed generation unit DG.

[0142] Coordination between DGC control systems needs to be achieved at the distribution network level, and coordination between DGC control systems and DG control systems needs to be achieved at the distributed generation cluster level. This is a coordination between a central coordinator and a distributed control unit.

[0143] In order to reduce information interaction between DGCs at the distribution network layer, a fully decentralized control method was designed. Within the DGC, the ADMM algorithm is used to decompose the calculation tasks of tracking the reactive power reference generated by the distribution network layer and minimizing the voltage deviation of nodes within the DGC into the DGC controller and the DG controller.

[0144] In the DGC control section of the distribution network layer, each DGC controller measures the voltage and reactive power injection information of its connected nodes and optimizes the calculation to obtain the optimal reactive power output reference using a fully decentralized solution method. The DGC layer control section updates global variables based on the reactive power reference calculated by the distribution network layer, combined with the objective of minimizing voltage deviations among nodes within the cluster, and transmits this information to the DG controller. The DG controller then updates its local and dual variables based on the updated global variables, thereby obtaining the DG output that minimizes voltage deviations between the distribution network layer and the DGC itself.

[0145] The system architecture of this embodiment reduces the communication requirements and, while achieving stable control of the active distribution network voltage, reduces the node voltage deviation within the DGC.

[0146] The specific implementation methods of each control system are as follows:

[0147] To solve the voltage control problem, the DGC control system obtains the global optimal solution through global gradient information. At the same time, to reduce communication loss and improve calculation speed, the system takes into account the characteristics of the distribution network structure and sets a special weight matrix to decouple the global gradient information. This allows the gradient information to be obtained from the local voltage and reactive power of the DGC, thereby achieving a completely decentralized calculation.

[0148] During this process, considering the actual operating conditions and the internal DG state of the DGC, the reactive power output of the DGC is constrained, and the reactive power output is further optimized to approach the intermediate output value to improve the reactive power margin and meet the actual operating requirements. After calculation, the DGC control system updates the obtained optimal reactive power reference as a constraint into the global variable and passes it to the DG control system.

[0149] In other words, the operating state of the DGC control optimization method is as follows: the distribution network is actively running, and then the DGC obtains the voltage information, real-time reactive power information and reactive power output limit information transmitted from the DG control system of the connected nodes; then it calculates the DGC reactive power output reference that meets the overall voltage stability of the distribution network on-site, updates the global variables and feeds them back to the DG controller.

[0150] To address the voltage control problem within the DGC (Distribution Generation Control Unit), a Lagrangian function is established, and the ADMM (Advanced Dynamics Model) algorithm is used to decompose the problem. Simultaneously, to take into account the voltage control of the distribution network layer, the reactive power reference obtained by the DGC controller is used as a constraint on the sum of DG outputs. Based on this objective and constraint, the optimal reactive power output of each DG unit is obtained.

[0151] During this process, the DG controller updates its local and dual variables based on the global variables updated by the DGC controller, and then transmits the updated values ​​to the DGC controller. After several iterations, the reactive power control command is issued to the DG unit.

[0152] In other words, the operation of the DG control optimization method is as follows: the global variables of the DGC are obtained and updated, and then the DG control obtains the voltage information, real-time reactive power information and reactive power output limit information of the connected nodes; then the local variables and dual variables are updated and fed back to the DGC controller; after several iterations, the optimal reactive power output command of the DG unit is obtained and issued to each DG, so as to realize voltage control that takes into account both the distribution network layer and the internal DGC.

[0153] Specifically, the distribution network layer DGC control system includes:

[0154] The data acquisition module is configured to acquire the real-time voltage and real-time reactive power injection at the DGC access point based on the objective function and constraints of the distribution network layer control.

[0155] The first parameter value preset module is set to the upper and lower limits of reactive power constraints and the initial values ​​of global variables determined based on the reactive power output transmitted by the DG controller; specifically, it determines the reactive power limit constraints and network power flow constraints imposed on the DGC.

[0156] The DGC calculation and control module is configured to calculate the optimal reactive power compensation amount for the active distribution network based on the acquired active distribution network measurements using a distributed gradient projection algorithm. This optimal reactive power compensation amount is then fed back to the lower-level DG control system to track changes in the active distribution network node voltage in real time, thereby achieving voltage stability in the active distribution network. The specific solution method is the same as in Example 1 and will not be repeated here.

[0157] Furthermore, it also includes a storage module for data interaction and storage with the lower-level DG control system.

[0158] Specifically, the DG control system at the DGC layer includes:

[0159] The local voltage reactive power acquisition module is configured to acquire the real-time voltage at the DG access point and the real-time reactive power output of the DG unit according to the objective function and constraints of the distributed generation cluster layer control.

[0160] The second parameter value preset module is set to obtain the numerical values ​​of the constraints and the initial values ​​of local and dual variables. Specifically, it determines the reactive power output constraints and network power flow constraints imposed on each distributed generation unit.

[0161] The DG calculation and control module is configured to calculate local and dual variables using the Alternating Directional Multiplier Method (ADMM) based on the global variables updated by the DGC controller, in order to obtain the optimal DG reactive power output. The optimized result is then sent to the DG unit to control its reactive power output, thereby achieving voltage stability at the nodes within the distributed generation cluster. The specific solution method is the same as in Example 1 and will not be repeated here.

[0162] The actual DG equipment in the distribution network DGC layer, that is, the distributed generation unit in the actual active distribution network, is currently mainly photovoltaic power generation unit.

[0163] Furthermore, it also includes a bidirectional communication channel between the DGC control system and the DG control system at the DGC layer, establishing a bidirectional connection between the DGC control system and the DG control system, transmitting the updated global variables of the DGC control system to each DG controller, and then transmitting the updated local variables and dual variables of each DG controller to the DGC controller, thereby achieving the purpose of coordinated control of the voltage at each DG terminal in the DGC layer.

[0164] The distributed generation cluster layer considers the specific topology of the distributed generation cluster and optimizes the reactive power distribution of each DG unit within the cluster. The upper-level DGC controller obtains the overall reactive power reference of the cluster and updates the global optimization variables that minimize the voltage deviation of each node within the lower-level distributed generation cluster. This is then passed to the DG controller, which subsequently updates its local and dual variables to obtain the reactive power output of each DG unit that can further adjust the voltage of nodes within the DGC. The system and control method in this embodiment are more suitable for the existing distribution network structure, meaning that the distributed generation cluster operator can uniformly control the output of each distributed generation unit within it, and different control operators do not share information with each other.

[0165] The above description is merely a preferred embodiment of this disclosure and is not intended to limit this disclosure. Various modifications and variations can be made to this disclosure by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

[0166] While the specific embodiments of this disclosure have been described above in conjunction with the accompanying drawings, this is not intended to limit the scope of protection of this disclosure. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art without creative effort based on the technical solutions of this disclosure are still within the scope of protection of this disclosure.

Claims

1. A two-layer voltage control method for active distribution networks considering DG cluster access, characterized in that: It includes two control layers: the distribution network layer and the distributed generation cluster layer; In the upper-level distribution network layer control, the reactive power output of the DGC in the active distribution network is taken as the control object. An objective function is constructed that includes the voltage stability requirements of each node in the distribution network layer and the reactive power margin requirements of the DGC. The optimal reactive power output of the DGC is obtained by solving the function, and distributed real-time control is performed on each DGC. In the lower-level distributed generation cluster layer control, an objective function for the voltage stability requirements of the end nodes of the DG units within the DGC is constructed. Based on the reactive power output reference of the DGC obtained from the distribution network layer, the optimal reactive power output of each DG unit is obtained, and the DG units are controlled in a distributed manner. After constructing the objective function in the lower-level distributed generation cluster control steps, the following process is executed: Determine the reactive power output constraints and network power flow constraints imposed on each distributed generation unit; Based on the objective function and constraints, determine the required DGC network data measurements for the solution; The objective function is solved using the alternating direction multiplier method based on the measurement data to obtain the optimal reactive power output of each DG; The optimized calculation results are sent to the DG unit to achieve voltage stability control that takes into account both the distribution network layer and the distributed generation cluster layer. In the distribution network layer control, specifically the DGC control step, after constructing the objective function, the following process is executed: Determine the reactive power constraints and network flow constraints imposed on DGC. Based on the objective function and constraints, obtain the real-time voltage and real-time reactive power injection at the DGC access point; The optimal reactive power compensation amount for DGC is calculated based on the obtained active distribution network measurement values ​​using a distributed gradient projection algorithm. The calculation results are transmitted to the lower-level distributed generation cluster control for further calculation and allocation, and the voltage fluctuation of the active distribution network is tracked in real time to achieve voltage stability of the active distribution network.

2. The active distribution network two-layer voltage control method considering DG cluster access as described in claim 1, characterized in that, The process of distributed control of the DG unit includes: Construct the corresponding Lagrangian function based on the objective function of the distributed generation cluster layer control; The DGC controller updates global variables and passes them to the DG controller under the reactive power reference constraints obtained from the upper distribution network layer control calculation. The DG controller updates local variables based on the updated global variables and the reactive power output inequality constraints, then updates dual variables, and passes the updated variables to the DGC controller. Through the interaction between the DG controller and the DGC controller, the optimal reactive power output of the DG is obtained, thereby optimizing the voltage of each node within the DGC.

3. The active distribution network two-layer voltage control method considering DG cluster access as described in claim 1, characterized in that, The objective function for distributed generation cluster layer control is to minimize the sum of the squared differences between the voltage of each DG terminal node and its corresponding standard value. The constraints of distributed generation cluster layer control include: the reactive power output of each DG should be between the upper and lower limits of its capacity, and the sum of the reactive power output of all DGs should meet the reactive power requirements of the DGC, as well as the voltage constraint on the relationship between the reactive power of the DGC network and the voltage, expressed in terms of voltage sensitivity.

4. The active distribution network two-layer voltage control method considering DG cluster access as described in claim 1, characterized in that, The reactive power margin requirement of DGC in the control objective function of the distribution network layer is: the reactive power output of the distributed generation cluster in the active distribution network is close to the midpoint of the upper and lower limits of reactive power output. The voltage stability requirement in the control objective function of the distribution network layer is to minimize the difference between the voltage fluctuation of the distribution network nodes and the reference value.

5. The active distribution network two-layer voltage control method considering DG cluster access as described in claim 1, characterized in that... : Distributed real-time control is performed on each DGC, including: Based on the overall optimal objective of the active distribution network, the optimal solution is sought in the direction of the negative gradient. By scaling the gradient values ​​and setting the weight factor to a constant, the solution process is decoupled to be related only to local information measurements, thereby solving for the optimal solution of DG reactive power output.

6. The active distribution network two-layer voltage control method considering DG cluster access as described in claim 5, characterized in that, The gradient scaling method is as follows: a positive definite scaling matrix is ​​added as the gradient coefficient in the solution, and the positive definite scaling matrix is ​​approximated diagonally by the Hessian matrix.

7. A two-layer voltage control system for an active distribution network considering DG cluster access, characterized in that, The system includes a DGC control system and a DG control system at the distribution network layer. The DGC control system is configured in the DGC controller to implement the control steps of the distribution network layer in the active distribution network dual-layer voltage control method considering DG cluster access as described in any one of claims 1-6. The DG control system is configured in the DG controller to execute the control steps of the distributed generation cluster layer in the active distribution network dual-layer voltage control method considering DG cluster access as described in any one of claims 1-6, thereby realizing the control of the distributed generation unit (DG).

8. The active distribution network two-layer voltage control system considering DG cluster access as described in claim 7, characterized in that: The DGC control system includes: The data acquisition module is configured to acquire the real-time voltage and real-time reactive power injection at the DGC access point based on the objective function and constraints of the distribution network layer control. The first parameter value preset module is set to the upper and lower limits of reactive power constraints and the initial values ​​of global variables determined based on the reactive power output transmitted by the DG controller. The DGC calculation and control module is configured to calculate the optimal reactive power compensation amount of DGC based on the acquired active distribution network measurement values ​​using a distributed gradient projection algorithm, and then feed the obtained optimal reactive power compensation amount of DGC back to the lower-level DG control system. Alternatively, the DG control system includes: The local voltage reactive power acquisition module is configured to acquire the real-time voltage at the DG access point and the real-time reactive power output of the DG unit according to the objective function and constraints of the distributed generation cluster layer control. The second parameter value preset module is set to obtain the numerical values ​​of the constraints and the initial values ​​of local variables and dual variables; The DG calculation and control module is configured to calculate local and dual variables using the alternating direction multiplier method based on the global variables updated by the DGC controller, in order to obtain the optimal DG reactive power output, and then send the optimization results to the DG unit to control the reactive power output.

Images