Voltage control method and device of network-forming converter, storage medium and equipment

Abstract

The invention discloses a voltage control method and device of a network construction type converter, a storage medium and equipment. The method comprises the following steps: constructing an operation cost optimization function of a long-term optimization layer of the power distribution network; based on the photovoltaic output prediction curve and the load prediction curve of each node in the power distribution network, solving an operation cost optimization function, and obtaining a voltage reference value of each network-forming converter in a future preset time period; according to the local voltage of each network construction type converter and the voltage of other converters adjacent to each network construction type converter, respectively adjusting the droop coefficient of each network construction type converter in a middle-term balance layer of the power distribution network; constructing a droop control equation of a real-time compensation layer of the power distribution network; and solving a droop control equation based on the voltage reference value of each grid-forming converter and the adjusted droop coefficient to obtain reactive power of each grid-forming converter so as to suppress voltage flicker and oscillation. According to the invention, voltage fluctuation can be effectively suppressed below a threshold value, and voltage out-of-limit is avoided.

Description

Technical Field

[0001] This application relates to the field of voltage control technology for power systems, and in particular to a voltage control method, apparatus, storage medium, and equipment for a grid-type converter. Background Technology

[0002] In traditional distribution networks, power flows from high-voltage levels to low-voltage levels, with voltage decreasing monotonically along the feeders. However, with large-scale integration of distributed photovoltaic (PV) systems, especially during peak PV generation periods at midday when local loads are low, feeder power flows may reverse. The active power injected by PV far exceeds the local load absorption capacity, causing power flow to be fed back to the upstream grid, resulting in a significant voltage rise along the feeders, particularly at the end nodes. When PV penetration is extremely high, node voltages can easily exceed safe operating limits, a problem known as voltage over-limit.

[0003] Currently, a medium- to long-term centralized optimization approach is typically used to centrally optimize regulation resources across the entire network or region. However, this approach heavily relies on communication, and communication latency and reliability issues make it difficult to cope with sudden, short-term voltage disturbances. Summary of the Invention

[0004] In view of this, this application provides a voltage control method, device, storage medium and equipment for a grid-type converter, which mainly enables voltage fluctuations to be effectively suppressed below the threshold and avoids voltage exceeding the limit.

[0005] According to a first aspect of this application, a voltage control method for a grid-type converter is provided, the method comprising: Construct the operating cost optimization function for the long-term optimization layer of the power distribution network; Based on the photovoltaic output prediction curve and the load prediction curve of each node in the distribution network, the operating cost optimization function is solved to obtain the voltage reference value of each grid-type converter in the future preset time period. Based on the local voltage of each grid-type converter and the voltage of other converters adjacent to each grid-type converter, the droop coefficient of each grid-type converter is adjusted in the medium-term balancing layer of the distribution network. Construct the droop control equations for the real-time compensation layer of the power distribution network; Based on the voltage reference values ​​and adjusted droop coefficients of each grid-type converter, the droop control equations are solved to obtain the reactive power of each grid-type converter. The execution cycle of the long-term optimization layer is longer than that of the medium-term balancing layer, and the execution cycle of the medium-term balancing layer is longer than that of the real-time compensation layer. Each grid-type converter is controlled to output corresponding reactive power to suppress voltage flicker and oscillation.

[0006] According to a second aspect of this application, a voltage control device for a grid-type converter is provided, the device comprising: The first building unit is used to construct the operating cost optimization function of the long-term optimization layer of the distribution network; The first solution unit is used to solve the operating cost optimization function based on the photovoltaic output prediction curve and the load prediction curve of each node in the distribution network, and to obtain the voltage reference value of each grid-type converter in the future preset time period. The adjustment unit is used to adjust the droop coefficient of each grid-type converter in the medium-term balancing layer of the distribution network according to the local voltage of each grid-type converter and the voltage of other converters adjacent to each grid-type converter. The second construction unit is used to construct the droop control equations of the real-time compensation layer of the power distribution network; The second solution unit is used to solve the droop control equation based on the voltage reference value and the adjusted droop coefficient of each grid-type converter to obtain the reactive power of each grid-type converter. The execution cycle of the long-term optimization layer is longer than the execution cycle of the medium-term balancing layer, and the execution cycle of the medium-term balancing layer is longer than the execution cycle of the real-time compensation layer. The control unit is used to control the output of the corresponding reactive power of each grid-type converter to suppress voltage flicker and oscillation.

[0007] According to a third aspect of this application, a storage medium is provided that stores a computer program thereon, which, when executed by a processor, implements the voltage control method for the grid-type converter described above.

[0008] According to a fourth aspect of this application, an electronic device is provided, including a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor, wherein the processor executes the program to implement the voltage control method for the grid-type converter described above.

[0009] By employing the above technical solutions, the voltage control method, apparatus, storage medium, and equipment for a grid-connected converter provided in this application, compared with existing technologies, can decompose the voltage control task into three control layers with different time scales: long-term, medium-term, and real-time. Each layer operates independently yet is coupled with information, jointly achieving precise, stable, and cost-optimized control of the grid voltage. The real-time compensation layer of this application performs instantaneous reactive power compensation based on the adjusted droop coefficient, effectively suppressing voltage flicker and transient overvoltage, thereby solving the voltage over-limit problem. Simultaneously, the medium-term balancing layer, through interaction with key status information from adjacent converters, can adaptively adjust the droop coefficients of each grid-connected converter online, thereby balancing the reactive power distribution among multiple converters and suppressing circulating currents that may arise from parallel operation, ensuring coordinated action of regional regulation resources and avoiding the blindness of purely local control. Furthermore, the long-term optimization layer performs global optimization calculations based on predicted photovoltaic output and load data for a future period, ensuring that the system operates safely and economically over a longer period. Furthermore, this layered architecture decomposes the complex global optimization problem. Only the intermediate balancing layer requires low-speed, low-bandwidth neighbor communication, while the fastest-responding real-time compensation layer has no communication requirements at all. This design greatly reduces the dependence on communication infrastructure and avoids the single point of failure risk and communication delay bottleneck of centralized control.

[0010] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description

[0011] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings: Figure 1 A schematic flowchart of a voltage control method for a grid converter provided in an embodiment of this application is shown; Figure 2 This paper illustrates a flowchart of a method for adjusting the sag coefficient of a mid-term equalization layer according to an embodiment of this application. Figure 3 This paper illustrates the spatiotemporal voltage distribution of a high-proportion photovoltaic distribution network provided in an embodiment of this application. Figure 4 This paper shows the spatiotemporal distribution diagram of voltage in a high-proportion photovoltaic distribution network after adopting this solution, as provided in an embodiment of this application. Figure 5 A schematic diagram illustrating the suppression effect of the real-time compensation layer provided in this application embodiment on short-term voltage fluctuations is shown. Figure 6 A schematic diagram of the voltage control device for a grid-type converter provided in an embodiment of this application is shown. Detailed Implementation

[0012] The present application will be described in detail below with reference to the accompanying drawings and embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in the embodiments of the present application can be combined with each other.

[0013] Existing technologies rely heavily on communication, and communication delays and reliability issues make them ill-equipped to handle sudden, short-term voltage disturbances.

[0014] To address the aforementioned issues, this invention provides a voltage control method for grid-connected converters. Its main objectives include: constructing a three-tiered, hierarchical collaborative control architecture of "long-term optimization - mid-term balancing - real-time compensation," decoupling the complex voltage control problem across different time scales to achieve minute-level cost optimization settings, second-level regional reactive power balancing, and millisecond-level instantaneous voltage disturbance suppression; designing a dynamic adaptive adjustment mechanism for the droop coefficient, enabling online adjustment through low-speed communication between neighboring units to eliminate reactive power circulation during multi-unit parallel operation, thereby achieving on-demand and balanced reactive power allocation; and proposing a cross-time-scale resource collaboration strategy, combining the day-ahead or intraday cost optimization scheduling results (such as voltage reference values ​​and energy storage charging / discharging plans) of the long-term optimization layer with the dynamic adjustments of the mid-term balancing layer and the real-time compensation layer, collaboratively utilizing the reactive power regulation capability of the converter and the active power support capability of energy storage to efficiently solve voltage limit exceedance and fluctuation problems under high-proportion photovoltaic access.

[0015] Based on this, this invention innovatively proposes a three-level collaborative control architecture, decomposing the voltage control task into three control layers with different time scales: long-term, medium-term, and real-time. Each layer has different control objectives, response times, and information requirements. Each layer operates independently yet is coupled with information, jointly achieving precise, stable, and economical control of the grid voltage. The overall system architecture comprises three collaborative layers, each with different control objectives, response times, and information requirements. The long-term optimization layer operates on a timescale of minutes to hours, acting as a "global planner." Its primary objective is cost optimization, i.e., performing global optimization calculations based on predicted photovoltaic output and load data for a future period. This layer does not pursue rapid response but rather provides optimized operating setpoints and boundary conditions for the medium-term balancing layer and the real-time compensation layer, thereby ensuring the system can operate safely and economically over a longer period. The medium-term balancing layer operates on a timescale of seconds, acting as a "regional coordinator." Its main purpose is to evenly distribute reactive power among multiple grid-connected converters and suppress circulating currents that may arise from parallel operation. It communicates with neighboring converters via a low-bandwidth sparse communication network. The layers exchange key state information (such as average voltage, reactive power output, etc.) and, based on consensus protocols or distributed algorithms, adaptively adjust their respective droop control systems online. This layer ensures coordinated action of regulation resources within the region and avoids the blindness of purely local control. The real-time compensation layer operates at the millisecond level and is the first line of defense for voltage control, playing the role of a "rapid response unit." Its goal is to quickly suppress instantaneous voltage disturbances caused by load changes, faults, or drastic fluctuations in photovoltaic output. This layer is entirely based on locally measured voltage and current information. Through the adjusted droop coefficient, it achieves rapid correction of the voltage reference value and instantaneous reactive power compensation, thereby effectively suppressing voltage flicker and transient overvoltage.

[0016] Furthermore, the embodiments of the invention provide a voltage control method for a grid-type converter, such as... Figure 1 As shown, the method includes: Step 10: Construct the operating cost optimization function for the long-term optimization layer of the power distribution network.

[0017] In this embodiment of the invention, when constructing the operating cost optimization function for the long-term optimization layer, a network loss cost function, a regulation cost function, and a voltage deviation penalty function for the distribution network are constructed respectively. The voltage deviation penalty function is multiplied by the penalty weight and then added to the network loss cost function and the regulation cost function in sequence to obtain the operating cost optimization function. The network loss cost function is used to calculate the electricity cost corresponding to the active power loss (network loss) of the distribution network lines throughout the day. The regulation cost function is the sum of the loss / lifetime depreciation cost function of energy storage active power charging and discharging, the equivalent cost function of reactive power provided by the converter, and the switching operation cost function of traditional capacitor banks. For the voltage deviation penalty function, in order to ensure voltage quality, this embodiment of the invention includes the degree to which the node voltage deviates from its nominal value as a penalty term in the objective function to guide the optimization result to make the voltage as close as possible to the ideal value.

[0018] The embodiments of the present invention construct a multi-objective optimization model, namely the operation cost optimization function, based on the network loss cost function, the regulation cost function, and the voltage deviation penalty function.

[0019] Step 20: Based on the photovoltaic output prediction curve and the load prediction curve of each node in the distribution network, solve the operating cost optimization function to obtain the voltage reference value of each grid-type converter in the future preset time period.

[0020] In this embodiment of the invention, when solving the operating cost optimization function, the photovoltaic output prediction curve, the load prediction curve of each node in the distribution network, the initial state of charge and charging / discharging efficiency of each energy storage unit, the grid topology and line parameters, as well as the electricity price information and equipment adjustment costs are first obtained, and the constraints of the operating cost optimization function are determined. Then, based on the photovoltaic output prediction curve, the load prediction curve of each node in the distribution network, the initial state of charge and charging / discharging efficiency of each energy storage unit, the grid topology and line parameters, the electricity price information and equipment adjustment costs, and the constraints, the operating cost optimization function is solved to obtain the voltage reference value and basic reactive power output plan of each grid-type converter in the future preset time period, the active power charging / discharging plan of each energy storage unit, and the switching status of the traditional capacitor composition.

[0021] Each node includes substations and loads, as well as all nodes in the distribution network where voltage needs to be considered. The preset time period can be set according to actual business needs; this embodiment of the invention does not impose specific limitations on this, such as setting the voltage reference values ​​for each grid-type converter for the next day. The photovoltaic output prediction curve and the load prediction curves for each node in the distribution network are the prediction data for the preset time period.

[0022] Specifically, the input data for the line cost optimization function mainly includes: photovoltaic output forecast curves, load forecast curves for each node in the distribution network, initial state of charge and charge / discharge efficiency of each energy storage unit, grid topology and line parameters, as well as electricity price information and equipment regulation costs. In addition, the constraints include: power flow constraints, security constraints, converter capacity constraints, and energy storage operation constraints. Among them, the power flow constraint condition refers to the requirement to satisfy the power flow equation of the distribution network (such as the DistFlow model and its second-order cone relaxation form); the safety constraint condition refers to the requirement that the voltage of all nodes must be within a safe range (such as [0.93, 1.07] pu) and the current of all branches cannot exceed their thermal stability limit; the converter capacity constraint condition refers to the requirement that the active and reactive power output of photovoltaic and energy storage converters must be within their apparent power circle (P² + Q² ≤ S²). In order to reserve adjustment margin for the medium-term balancing layer and the real-time compensation layer, the reactive power plan issued by the long-term optimization layer is usually within a small range (such as ±30% of the rated power); the energy storage operation constraint condition refers to the requirement that the SOC of energy storage must be between the allowable upper and lower limits, the charging and discharging power cannot exceed the rated value, and the final SOC of energy storage should be restored to the initial state at the end of a complete scheduling cycle to ensure its sustainable operation.

[0023] During execution, the long-term optimization layer employs a rolling optimization approach. For example, every 15 minutes or 1 hour, the operating cost optimization function is re-solved based on the latest forecast data. Its output decision variables include: the active power charging and discharging plans of each energy storage unit over a future period, the voltage reference value (V_ref) and basic reactive power output plan of each grid-connected converter, and the switching status of traditional capacitor banks. Through command transmission, the distribution automation system distributes these decision variables to the medium-term balancing layer and the real-time compensation layer as their operating benchmarks.

[0024] Step 30: Based on the local voltage of each grid-type converter and the voltage of other converters adjacent to each grid-type converter, adjust the droop coefficient of each grid-type converter in the medium-term balancing layer of the distribution network.

[0025] In the embodiments of the present invention, the intermediate balancing layer is the key to realizing the coordination of multiple grid-type converters. It is dynamically fine-tuned within the framework set by the long-term optimization layer to solve the problem of "each fighting its own battle".

[0026] Regarding the coordinated operation process between various grid-type converters, and the droop coefficient adjustment process, such as... Figure 2 As shown, it includes: Step 31: Control each grid-type converter to interact with other adjacent converters to obtain the voltage and reactive power output of the other converters.

[0027] In this embodiment of the invention, the intermediate balancing layer does not require a high-speed, fully connected communication network. Instead, it uses low-speed communication (such as CAN bus, RS485, or wireless Zigbee) to allow each converter to interact with its physical or logical neighbors. The communication content is concise, typically containing only key information such as its own voltage and reactive power output. The communication frequency is approximately once every few seconds (e.g., 10 seconds), and the communication bandwidth requirement is extremely low (e.g., less than 1 kbps).

[0028] Step 32: For any one of the grid-type converters, perform a weighted average calculation based on the local voltage of the grid-type converter and the voltages of other adjacent converters to obtain the local average voltage.

[0029] Traditional droop coefficients are fixed values, leading to reactive power distribution errors due to line impedance differences. To address this, this invention innovatively proposes an adaptive droop coefficient adjustment method. Specifically, it employs a consensus protocol-based adjustment strategy: each converter i performs a weighted average of its local voltage and the voltage received from its neighbor j to form a local average voltage. Then, based on the deviation between its own voltage and this local average voltage, it adjusts its droop coefficient m_q online.

[0030] Step 33: Determine the deviation between the local voltage of any one of the grid-type converters and the local average voltage, and adjust the droop coefficient of any one of the grid-type converters based on the deviation to achieve reactive power balancing distribution.

[0031] Specifically, if the voltage V_i of a converter i is significantly higher than the average level of its neighbors, it indicates that it is undertaking an excessive reactive power support task (or absorbing too much reactive power). In this case, its droop coefficient m_q should be increased to make it more sluggish in response to voltage changes, thereby freeing up some reactive power regulation tasks for other units. Conversely, if the voltage V_i is too low, the droop coefficient m_q should be decreased to make it participate more actively in voltage regulation. Through this distributed, iterative adjustment, the droop coefficients of all participating converters will tend to converge, ultimately achieving a balanced distribution of reactive power according to converter capacity or other preset weights, effectively eliminating circulating current. The real-time compensation layer can dynamically adjust and output a time-varying droop coefficient m_q(t), which will be directly used by the real-time compensation layer in its droop control equation.

[0032] Step 40: Construct the droop control equations for the real-time compensation layer of the power distribution network.

[0033] In this embodiment of the invention, a basic droop control function and a local voltage change rate function are constructed for the distribution network; the local voltage change rate function is multiplied by the feedforward gain coefficient to obtain the feedforward compensation function; based on the basic droop control function and the feedforward compensation function, the droop control equation for the real-time compensation layer of the distribution network is constructed. The specific droop control equation is as follows:

[0034] Wherein, the basic droop control function is , It is a voltage reference value issued by the long-term optimization layer. This is the reactive power reference value, which is the set value. It is the droop coefficient for dynamic adjustment of the mid-term equilibrium layer. It is the instantaneous sample value of the local grid-connected voltage. Q Let be the reactive power of the grid-type converter, and be the parameter to be solved. The feedforward gain coefficient is a pre-set value. This represents the local voltage change rate.

[0035] Step 50: Based on the voltage reference values ​​and adjusted droop coefficients of each grid-type converter, solve the droop control equations to obtain the reactive power of each grid-type converter.

[0036] The execution cycle of the long-term optimization layer is longer than that of the intermediate-term balancing layer, and the execution cycle of the intermediate-term balancing layer is longer than that of the real-time compensation layer.

[0037] In the specific solution of this invention, the voltage reference value and the adjusted droop coefficient of each grid-type converter, as well as the instantaneous sampled value of the local grid-connected voltage, are substituted into the droop control equation for solution to obtain the reactive power of each grid-type converter.

[0038] Specifically, the feedforward compensation function This is an innovative aspect of the present invention: extracting the local voltage deviation through a high-pass filter or differentiating element. The high-frequency component (i.e., the rate of voltage change) is multiplied by a feedforward gain factor. When a disturbance in the power grid causes a rapid voltage drop or rise, this component quickly generates a compensating voltage component that moves in the opposite direction to the change. This is equivalent to injecting virtual inertia and damping into the system, enabling rapid intervention before the voltage deviates from its steady-state value, effectively suppressing voltage flicker and oscillations. The real-time compensation layer can respond within 10 milliseconds, providing rapid dynamic voltage support to the power grid and significantly improving the system's voltage recovery capability and stability in the face of large disturbances.

[0039] Step 60: Control each grid-type converter to output the corresponding reactive power to suppress voltage flicker and oscillation.

[0040] In this embodiment of the invention, the three control levels coordinate through information flow. The V_ref output by the long-term optimization layer serves as the control objective for the intermediate equalization layer and the implementation compensation layer. The droop coefficient m_q(t) output by the intermediate equalization layer is a key control parameter for the real-time compensation layer. This top-down instruction transmission ensures that the rapid control at the lower levels always serves the overall goal of cost optimization at the upper levels. Furthermore, to avoid conflicts and oscillations between control loops of different time scales, this embodiment employs a time constant separation design principle, using low-pass filters with different cutoff frequencies within the controllers of each level. For example, the filter time constant of the real-time compensation layer is approximately 5ms, the filter time constant of the intermediate equalization layer is approximately 10s, and the long-term optimization layer is at the minute level. This design ensures that each control layer only responds to disturbances in its corresponding frequency band, achieving effective decoupling of control at different time scales.

[0041] The embodiments of the present invention have been verified through simulation, and their performance has been compared with that of traditional methods, as shown in Table 1.

[0042] Compared with traditional methods, the multi-time-scale voltage collaborative control method proposed in this invention has the following significant advantages: This invention employs a three-level collaborative control system to handle voltage disturbances ranging from milliseconds to hours. The real-time compensation layer effectively suppresses instantaneous voltage flicker and drops, the mid-term balancing layer addresses circulating current and reactive power imbalance issues in multi-machine parallel operation, and the long-term optimization layer ensures the system's voltage qualification rate under various operating conditions. Simulation results demonstrate that, under the same disturbance, this invention reduces the voltage recovery time from over 500ms using traditional methods to within 200ms, and significantly reduces the steady-state voltage deviation from ±3% to within ±0.8%, representing an improvement of over 70%.

[0043] Meanwhile, this invention, through a long-term optimization layer, comprehensively considers network losses, regulation costs, and equipment lifespan, achieving optimized scheduling of various resources such as energy storage and converters. It avoids increased network losses due to improper reactive power compensation and reduces peak electricity purchase costs through the "peak shaving and valley filling" effect of energy storage. Compared to traditional control strategies that only consider voltage safety, this invention can reduce the total daily operating cost of the system by 5%-10%.

[0044] Furthermore, this embodiment of the invention employs a layered distributed architecture, decomposing the complex global optimization problem. Only the intermediate balancing layer requires low-speed, low-bandwidth proximity communication, while the fastest-responding real-time layer has no communication requirements whatsoever. This design of this embodiment of the invention significantly reduces dependence on communication infrastructure, avoiding the single point of failure risk and communication latency bottlenecks inherent in centralized control.

[0045] Furthermore, this embodiment of the invention, through an adaptive adjustment mechanism of the droop coefficient of the intermediate balancing layer, can effectively solve the problem of uneven reactive power output of multiple converters caused by line impedance imbalance. Simulation results show that this embodiment of the invention can reduce the reactive power imbalance of multiple units from 15%-20% under traditional fixed droop control to more than 5%, thereby ensuring coordinated operation of each unit and avoiding overload of individual units.

[0046] To make the technical solutions and beneficial effects of the embodiments of the present invention clearer, they will be described in detail below with reference to the accompanying drawings. Figure 3 This shows the voltage distribution of a typical distribution network over a 24-hour period without any advanced control measures. It is evident that during the midday period (approximately 10:00-15:00), due to high photovoltaic power generation, the voltage (yellow area) at multiple nodes (especially at the feeder ends) significantly exceeds the safe limit (e.g., 1.07 pu), indicating a serious voltage exceedance problem.

[0047] Figure 4 The voltage distribution of the same distribution network after applying the control method described in the embodiments of the present invention is shown. It can be seen that the voltage of all nodes (mainly green and light blue) is effectively controlled within a safe range (e.g., [0.93, 1.07] pu) throughout the day, and the problem of voltage exceeding the upper limit during the midday period is completely solved. This proves the effectiveness of the embodiments of the present invention on a macroscopic time scale.

[0048] Figure 5 This displays the dynamic changes in the voltage of a node over a short timescale of minutes. The red curve (before control) shows that due to rapid fluctuations in photovoltaic output, the voltage repeatedly touched and exceeded the preset action threshold. The blue curve (after control) shows that the real-time compensation layer of this embodiment can respond quickly, effectively suppressing voltage fluctuations below the threshold through instantaneous reactive power adjustment, thus avoiding voltage exceeding the limit. This demonstrates the rapid response capability of this embodiment on a microscopic timescale.

[0049] like Figure 3 and Figure 4 As shown, after implementing this invention, the voltage qualification rate of the power distribution network increased from 85% before control to over 99.9% throughout the day, completely eliminating the voltage over-limit phenomenon.

[0050] like Figure 5As shown, in a scenario where cloud cover causes a 50% drop in photovoltaic output within 10 seconds, node voltage fluctuations are suppressed to within 1.5%, while fluctuations can reach 5% under traditional control.

[0051] By optimizing energy storage for charging during off-peak hours and discharging during peak hours, and reducing reactive power flow across the entire network, the total daily operating cost of the system has been reduced by approximately 8%.

[0052] The embodiments of the present invention do not require large-scale hardware modifications. They can significantly improve the acceptance capacity and operational efficiency of high-proportion renewable energy distribution networks by mainly upgrading the control strategy, and have high promotion value.

[0053] Furthermore, as Figures 1 to 2 The specific implementation of the method shown in this embodiment provides a voltage control device for a grid-type converter, such as... Figure 6 As shown, the device includes: a first construction unit 101, a first solution unit 102, an adjustment unit 103, a second construction unit 104, a second solution unit 105, and a control unit 106.

[0054] The first building unit 101 can be used to build the operating cost optimization function of the long-term optimization layer of the power distribution network.

[0055] The first solving unit 102 can be used to solve the operating cost optimization function based on the photovoltaic output prediction curve and the load prediction curve of each node in the distribution network, so as to obtain the voltage reference value of each grid-type converter in the future preset time period.

[0056] The adjustment unit 103 can be used to adjust the droop coefficient of each grid-type converter in the medium-term balancing layer of the distribution network according to the local voltage of each grid-type converter and the voltage of other converters adjacent to each grid-type converter.

[0057] The second building unit 104 can be used to build the droop control equations of the real-time compensation layer of the power distribution network.

[0058] The second solving unit 105 can be used to solve the droop control equation based on the voltage reference value and the adjusted droop coefficient of each grid-type converter to obtain the reactive power of each grid-type converter. The execution cycle of the long-term optimization layer is longer than the execution cycle of the medium-term balancing layer, and the execution cycle of the medium-term balancing layer is longer than the execution cycle of the real-time compensation layer.

[0059] The control unit 106 can be used to control the output of the corresponding reactive power of each grid-type converter to suppress voltage flicker and oscillation.

[0060] In some embodiments, the first construction unit 101 may be specifically used to construct the network loss cost function, the regulation cost function, and the voltage deviation penalty function of the distribution network respectively; after multiplying the voltage deviation penalty function by the penalty weight, it is added to the network loss cost function and the regulation cost function in sequence to obtain the operating cost optimization function.

[0061] In some embodiments, the first solving unit 102 may be specifically used to determine the constraints of the operating cost optimization function, the constraints including: power flow constraints, security constraints, converter capacity constraints, and energy storage operation constraints; based on the photovoltaic output prediction curve, the load prediction curve of each node in the distribution network, the initial state of charge and charging / discharging efficiency of each energy storage unit, the grid topology and line parameters, electricity price information and equipment adjustment costs, and the constraints, to solve the operating cost optimization function, and obtain the voltage reference value and basic reactive power output plan of each grid-type converter in the future preset time period, the active power charging / discharging plan of each energy storage unit, and the switching status of the traditional capacitor composition.

[0062] In some embodiments, the adjustment unit 103 may be specifically used to perform a weighted average calculation on any one of the grid-type converters, based on the local voltage of the grid-type converter and the voltages of other adjacent converters, to obtain a local average voltage; determine the deviation between the local voltage of the grid-type converter and the local average voltage; and adjust the droop coefficient of the grid-type converter based on the deviation to achieve reactive power balancing distribution.

[0063] In some embodiments, the control unit 106 can also be used to control each grid-type converter to interact with other adjacent converters to obtain the voltage and reactive power output of the other converters.

[0064] In some embodiments, the second construction unit 104 may be specifically used to construct the basic droop control function and the local voltage change rate function of the distribution network; multiply the local voltage change rate function by the feedforward gain coefficient to obtain the feedforward compensation function; and construct the droop control equation of the real-time compensation layer of the distribution network based on the basic droop control function and the feedforward compensation function.

[0065] In some embodiments, the second solving unit 105 may be specifically used to substitute the voltage reference value and adjusted droop coefficient of each grid-connected converter, as well as the instantaneous sampled value of the local grid-connected voltage, into the droop control equation for solving, so as to obtain the reactive power of each grid-connected converter.

[0066] It should be noted that other corresponding descriptions of the functional units involved in the voltage control device of the grid converter provided in this embodiment can be found in the following references. Figures 1 to 2 The corresponding descriptions in [the document] will not be repeated here.

[0067] Based on the above, Figures 1 to 2 Accordingly, this embodiment also provides a storage medium storing a computer program that, when executed by a processor, implements the above-described method. Figures 1 to 2 The voltage control method of the grid-type converter is shown.

[0068] Based on this understanding, the technical solution of this application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, portable hard drive, etc.) and includes several instructions to cause an electronic device (such as a personal computer, server, or network device, etc.) to execute the methods of various implementation scenarios of this application.

[0069] Based on the above, Figures 1 to 2 The method shown, and Figure 6 To achieve the above objectives, the present application also provides an electronic device, specifically a personal computer, tablet computer, server, or other network device, as shown in the virtual device embodiment. This device includes a storage medium and a processor; the storage medium stores a computer program; the processor executes the computer program to achieve the above-described objectives. Figures 1 to 2 The voltage control method of the grid-type converter is shown.

[0070] Optionally, the aforementioned physical devices may also include a user interface, a network interface, a camera, radio frequency (RF) circuitry, sensors, audio circuitry, a Wi-Fi module, etc. The user interface may include a display screen, input units such as a keyboard, etc., and optional user interfaces may also include USB interfaces, card reader interfaces, etc. The network interface may optionally include standard wired interfaces, wireless interfaces (such as Wi-Fi interfaces), etc.

[0071] Those skilled in the art will understand that the physical device structure provided in this embodiment does not constitute a limitation on the physical device, and may include more or fewer components, or combine certain components, or have different component arrangements.

[0072] The storage medium may also include an operating system and a network communication module. The operating system is a program that manages the hardware and software resources of the aforementioned physical device, supporting the operation of information processing programs and other software and / or programs. The network communication module is used to enable communication between the various components within the storage medium, as well as communication with other hardware and software in the information processing physical device.

[0073] Through the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented by means of software plus necessary general-purpose hardware platform, or it can be implemented by hardware.

[0074] This invention decomposes voltage control tasks into three control layers with different time scales: long-term, medium-term, and real-time. Each layer operates independently yet is coupled with information, jointly achieving precise, stable, and cost-optimized control of the grid voltage. The real-time compensation layer of this invention performs instantaneous reactive power compensation based on the adjusted droop coefficient, effectively suppressing voltage flicker and transient overvoltage, thus solving voltage limit exceedance problems. Simultaneously, the medium-term balancing layer, through interaction with adjacent converters regarding key status information, can adaptively adjust the droop coefficients of each grid-connected converter online, thereby evenly distributing reactive power among multiple converters and suppressing circulating currents that may arise from parallel operation. This ensures coordinated operation of regional regulation resources and avoids the blindness of purely local control. Furthermore, the long-term optimization layer performs global optimization calculations based on predicted photovoltaic output and load data for a future period, ensuring both safe and economical system operation over a longer period. Furthermore, this layered architecture in the embodiments of the present invention decomposes the complex global optimization problem. Only the intermediate balancing layer requires low-speed, low-bandwidth neighbor communication, while the real-time compensation layer with the fastest response has no communication requirements at all. This design greatly reduces the dependence on communication infrastructure and avoids the single point of failure risk and communication delay bottleneck of centralized control.

[0075] Those skilled in the art will understand that the accompanying drawings are merely schematic diagrams of a preferred embodiment, and the modules or processes shown in the drawings are not necessarily essential for implementing this application. Those skilled in the art will understand that the modules in the apparatus of the embodiment can be distributed within the apparatus of the embodiment as described, or can be modified to be located in one or more apparatuses different from this embodiment. The modules of the above-described embodiment can be combined into one module, or further divided into multiple sub-modules.

[0076] The serial numbers in this application are for descriptive purposes only and do not represent the superiority or inferiority of any particular implementation scenario. The above disclosures are merely a few specific implementation scenarios of this application; however, this application is not limited thereto, and any variations conceived by those skilled in the art should fall within the protection scope of this application.

Claims

1. A voltage control method for a grid-type converter, characterized in that, include: Construct the operating cost optimization function for the long-term optimization layer of the power distribution network; Based on the photovoltaic output prediction curve and the load prediction curve of each node in the distribution network, the operating cost optimization function is solved to obtain the voltage reference value of each grid-type converter in the future preset time period. Based on the local voltage of each grid-type converter and the voltage of other converters adjacent to each grid-type converter, the droop coefficient of each grid-type converter is adjusted in the medium-term balancing layer of the distribution network. Construct the droop control equations for the real-time compensation layer of the power distribution network; Based on the voltage reference values ​​and adjusted droop coefficients of each grid-type converter, the droop control equations are solved to obtain the reactive power of each grid-type converter. The execution cycle of the long-term optimization layer is longer than that of the medium-term balancing layer, and the execution cycle of the medium-term balancing layer is longer than that of the real-time compensation layer. Each grid-type converter is controlled to output corresponding reactive power to suppress voltage flicker and oscillation.

2. The method according to claim 1, characterized in that, The operating cost optimization function for constructing the long-term optimization layer of the distribution network includes: Construct the network loss cost function, regulation cost function, and voltage deviation penalty function of the distribution network respectively; The voltage deviation penalty function is multiplied by the penalty weight and then added to the network loss cost function and the adjustment cost function in sequence to obtain the operating cost optimization function.

3. The method according to claim 1, characterized in that, Based on the photovoltaic output prediction curve and the load prediction curve of each node in the distribution network, the operating cost optimization function is solved to obtain the voltage reference values ​​of each grid-type converter within a preset future time period, including: The constraints of the operating cost optimization function are determined, including: power flow constraints, safety constraints, converter capacity constraints, and energy storage operation constraints. Based on the photovoltaic power output prediction curve, the load prediction curve of each node in the distribution network, the initial state of charge and charging / discharging efficiency of each energy storage unit, the grid topology and line parameters, electricity price information and equipment adjustment costs, and the constraints, the operating cost optimization function is solved to obtain the voltage reference value and basic reactive power output plan of each grid-type converter in the future preset time period, the active power charging / discharging plan of each energy storage unit, and the switching status of the traditional capacitor group.

4. The method according to claim 1, characterized in that, The step of adjusting the droop coefficient of each grid-type converter in the medium-term balancing layer of the distribution network based on the local voltage of each grid-type converter and the voltage of other converters adjacent to each grid-type converter includes: For any one of the grid-type converters, a weighted average calculation is performed based on the local voltage of the grid-type converter and the voltages of other adjacent converters to obtain the local average voltage. The deviation between the local voltage of any one of the grid-type converters and the local average voltage is determined, and the droop coefficient of any one of the grid-type converters is adjusted based on the deviation to achieve reactive power balancing distribution.

5. The method according to claim 4, characterized in that, The method further includes: Each grid-type converter is controlled to interact with its adjacent converters to obtain the voltage and reactive power output of the other converters.

6. The method according to claim 1, characterized in that, The droop control equations for constructing the real-time compensation layer of the distribution network include: Construct the basic droop control function and the local voltage change rate function of the power distribution network; Multiplying the local voltage change rate function by the feedforward gain coefficient yields the feedforward compensation function; Based on the basic droop control function and the feedforward compensation function, the droop control equations of the real-time compensation layer of the distribution network are constructed.

7. The method according to claim 1, characterized in that, The process involves solving the droop control equation based on the voltage reference values ​​and adjusted droop coefficients of each grid-connected converter to obtain the reactive power of each grid-connected converter, including: The voltage reference value and adjusted droop coefficient of each grid-connected converter, as well as the instantaneous sampled value of the local grid voltage, are substituted into the droop control equation for solution to obtain the reactive power of each grid-connected converter.

8. A voltage control device for a grid-type converter, characterized in that, include: The first building unit is used to construct the operating cost optimization function of the long-term optimization layer of the distribution network; The first solution unit is used to solve the operating cost optimization function based on the photovoltaic output prediction curve and the load prediction curve of each node in the distribution network, and to obtain the voltage reference value of each grid-type converter in the future preset time period. The adjustment unit is used to adjust the droop coefficient of each grid-type converter in the medium-term balancing layer of the distribution network according to the local voltage of each grid-type converter and the voltage of other converters adjacent to each grid-type converter. The second construction unit is used to construct the droop control equations of the real-time compensation layer of the power distribution network; The second solution unit is used to solve the droop control equation based on the voltage reference value and the adjusted droop coefficient of each grid-type converter to obtain the reactive power of each grid-type converter. The execution cycle of the long-term optimization layer is longer than the execution cycle of the medium-term balancing layer, and the execution cycle of the medium-term balancing layer is longer than the execution cycle of the real-time compensation layer. The control unit is used to control the output of the corresponding reactive power of each grid-type converter to suppress voltage flicker and oscillation.

9. A storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 7.

10. An electronic device comprising a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method of any one of claims 1 to 7.

Images