SVG adaptive switching control method and system based on operating short-circuit ratio

Abstract

This invention relates to the field of power system stability control technology, and discloses an adaptive switching control method and system for SVG (Static Var Generator) based on the operating short-circuit ratio and grid configuration. The method includes: real-time acquisition of voltage and current phasor data and renewable energy output data at the grid connection point; calculation of the real-time short-circuit ratio of the current system based on the ratio of the grid's inherent short-circuit capacity to the real-time output of renewable energy; calculation of the target number of SVGs operating in the required grid configuration mode using a piecewise function under constraints of minimum and maximum grid configuration ratios; comparison of the number of SVGs currently operating in the grid configuration mode with the target number, and selection of the target SVG to trigger the switching process according to preset rules; and switching the target SVG from the current control mode to the target control mode using a smooth transition algorithm, while coordinating the control parameters of the remaining SVGs. This invention improves the reliability of the control logic by combining multi-timescale short-circuit ratio calculation with validity verification.

Description

Technical Field

[0001] This invention relates to the field of power system stability control technology, specifically to an SVG-based adaptive switching control method and system for network structure based on the operating short-circuit ratio. Background Technology

[0002] With the large-scale integration of new energy sources such as photovoltaics and wind power into the power grid, the power system exhibits significant characteristics of high proportion of power electronics, low rotational inertia, and weak grid connections. The output of new energy sources is highly random and volatile due to natural conditions, directly leading to real-time dynamic changes in the short-circuit ratio (SCR) at the grid connection point: During peak periods of new energy output, the parallel operation of numerous units reduces the system's equivalent impedance, decreasing the SCR (SCR<2), resulting in a weak grid characteristic that is prone to voltage instability and subsynchronous oscillations; during off-peak periods of new energy output, the reduction in parallel units increases the system's equivalent impedance and the SCR (SCR>3), resulting in a strong grid characteristic, where excessive grid control can easily trigger high-frequency oscillations and control loop instability.

[0003] Static var generators (SVG), as core reactive power compensation and voltage support devices in power systems, have control strategies mainly divided into two categories: grid-following (GFL) and grid-forming (GFM). Grid-following control tracks the grid voltage phase through a phase-locked loop (PLL), exhibiting current source characteristics. It relies on grid voltage support and performs well under strong grid conditions (high SCR), but is prone to instability under weak grid conditions (low SCR) due to PLL bandwidth limitations. Grid-forming control does not require a PLL and can autonomously construct voltage amplitude and phase, exhibiting voltage source characteristics. It can provide inertia and damping to the grid, effectively providing voltage support under weak grid conditions, but is prone to oscillations due to over-adjustment under strong grid conditions.

[0004] Existing SVG control technology has several limitations: First, the control mode is fixed, and SVGs are mostly set to a single control mode after leaving the factory, which cannot adapt to the real-time changes in the short-circuit ratio of the power grid. Second, the short-circuit ratio calculation lacks multi-timescale design, relying only on offline or single real-time data, which is easily affected by transient disturbances and leads to distorted calculation results. Third, the decision on the number of SVGs in the grid configuration mode is coarse, without clear segmented quantitative calculation rules, and does not strictly follow the minimum / maximum grid configuration ratio constraints, which can easily lead to system instability due to full switching. Fourth, the mode switching lacks refined equipment selection rules and smooth transition mechanisms, and the switching process is prone to current surges and voltage fluctuations, and the time constraint for switching triggering is fixed, which cannot adapt to the trend changes or short-term disturbances of the power grid. Fifth, the multi-SVG coordinated control is insufficient. After the mode switch, the reactive power reference value and PLL bandwidth of the remaining grid-connected SVGs are not adjusted synchronously, resulting in poor overall compensation effect of the SVG cluster.

[0005] In existing related technologies, some solutions propose grid-connected / grid-following switching based on the short-circuit ratio, but these mostly focus on the renewable energy inverters themselves and do not involve the coordinated control of SVG clusters. Some solutions only optimize the single control strategy of grid-connected SVGs, failing to achieve adaptive grid-connected / grid-following switching. Other solutions lack quantitative calculation models and smooth transition designs for their switching logic, failing to meet the stability requirements of actual engineering projects. Therefore, there is an urgent need for an adaptive switching method that can sense changes in the grid short-circuit ratio in real time, quantitatively determine the number of grid-connected SVGs, smoothly complete mode switching, and achieve coordinated control of multiple SVGs, in order to adapt to the grid characteristics of a high proportion of renewable energy connected to the grid. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention provides an adaptive switching control method and system for SVG (Static Var Generator) based on the short-circuit ratio, achieving the following objectives: Accurate calculation of the short-circuit ratio across multiple time scales, combining real-time, trend, and predictive data for validity verification to avoid calculation distortion caused by transient disturbances; Establishing a quantitative calculation model for the target number of network SVGs based on piecewise functions, adapting to grid demands in different short-circuit ratio ranges under minimum / maximum network ratio constraints, and eliminating the extreme risk of full switching; Designing refined target SVG selection rules, combining equipment status, electrical distance, and historical operating time to achieve optimal switching and extend the overall lifespan of the SVG cluster; Proposing a three-stage smooth transition algorithm including pre-switching preparation to achieve shock-free switching of control modes, while designing dynamic time constraints to accurately distinguish between grid trend changes and short-term disturbances, avoiding ineffective switching; Achieving coordinated optimization of control parameters for the remaining SVGs after mode switching, synchronously adjusting the reactive power reference value and PLL bandwidth of the network-connected SVGs to ensure the overall reactive power compensation and grid support effect of the SVG cluster.

[0007] This invention is achieved through the following technical solution:

[0008] An adaptive switching control method for SVG trailing networks based on the short-circuit ratio includes:

[0009] Real-time collection of voltage and current phasor data and renewable energy output data at grid connection points; calculation of the real-time short-circuit ratio of the current system based on the ratio of the grid's inherent short-circuit capacity to the real-time output of renewable energy.

[0010] Based on the relationship between the real-time short-circuit ratio and the first short-circuit ratio threshold and the second short-circuit ratio threshold, under the constraints of the minimum network ratio and the maximum network ratio, the number of SVG targets running in the required network mode is calculated using a piecewise function, where the first short-circuit ratio threshold is less than the second short-circuit ratio threshold.

[0011] Compare the number of SVGs currently running in network construction mode with the target number. When the absolute value of the difference between the number of SVGs currently running in network construction mode and the target number exceeds a preset number threshold and meets a preset time constraint, determine whether to switch some network-following mode SVGs to network construction mode or switch some network-following mode SVGs to network-following mode based on the sign of the difference. Select the target SVG according to the preset rules and trigger the switching process.

[0012] The target SVG is switched from the current control mode to the target control mode using a smooth transition algorithm, and the control parameters of the remaining SVGs are coordinated.

[0013] As an optimization, the calculation of the real-time short-circuit ratio includes multiple time scales:

[0014] The real-time short-circuit ratio is calculated based on the data collected in real time by the phasor measurement unit, with the first update cycle.

[0015] The real-time short-circuit ratio is averaged, and the trend short-circuit ratio is calculated with the second update period;

[0016] Based on the new energy power output forecast data, the predicted short-circuit ratio is calculated in the third update cycle. The new energy power output forecast data includes at least photovoltaic power output forecast and wind power output forecast based on meteorological data.

[0017] The validity of the real-time short-circuit ratio is verified. When the difference between the real-time short-circuit ratio and the trend short-circuit ratio exceeds a preset deviation threshold, it is determined to be invalid data, and the previous valid value is maintained.

[0018] As an optimization, the calculation of the number of SVG targets for the required network configuration mode using a piecewise function specifically includes:

[0019] When the inherent short-circuit capacity of the power grid is less than the product of the first short-circuit ratio threshold and the real-time output of new energy, the product of the maximum grid construction ratio and the total number of SVG units is taken as the target number.

[0020] When the inherent short-circuit capacity of the power grid is greater than the product of the second short-circuit ratio threshold and the real-time output of new energy, the product of the minimum grid construction ratio and the total number of SVG units is taken as the target number.

[0021] When the inherent short-circuit capacity of the power grid is between the product of the first short-circuit ratio threshold and the real-time output of new energy sources, and the product of the second short-circuit ratio threshold and the real-time output of new energy sources, the target quantity is calculated using a theoretical formula, and then rounded after applying the constraints of the minimum and maximum grid construction ratios.

[0022] As an optimization, when the inherent short-circuit capacity of the power grid is between the product of the first short-circuit ratio threshold and the real-time output of new energy sources, and the second short-circuit ratio threshold and the real-time output of new energy sources, the target quantity is calculated using a theoretical formula:

[0023] ;

[0024] in, For the target short-circuit ratio, Rated capacity for a single SVG unit. For the network support coefficient, For the inherent short-circuit capacity of the power grid, Providing real-time power for new energy sources The theoretical computational cost of the SVG in the network configuration mode;

[0025] The target number after applying minimum and maximum meshing ratio constraints and rounding is:

[0026] ;

[0027] in, To achieve the maximum network ratio, To minimize the network ratio, This represents the total number of SVG units. The final number of SVG targets for the desired mesh pattern.

[0028] As an optimization, the target SVG is selected according to preset rules, specifically including:

[0029] A. Selection rules when deploying network configuration modes:

[0030] When it is necessary to switch some SVGs from mesh mode to mesh construction mode, the target SVGs are selected from the SVGs currently running in mesh mode according to the following priority order:

[0031] First priority: SVG with no fault alarms, no overheating, no communication abnormalities, and remaining capacity greater than the remaining capacity threshold;

[0032] Second priority: Among the SVGs that meet the first priority, SVGs with an electrical distance less than the first distance threshold are selected first, and the smaller the electrical distance, the higher the priority.

[0033] Third priority: When the predicted short-circuit ratio is continuously lower than the first short-circuit ratio threshold for a period of time exceeding the preset duration threshold, among the SVGs that meet the first and second priorities, the SVGs whose historical cumulative network construction running time is less than the first running time threshold are selected first, and the shorter the historical running time, the higher the priority.

[0034] B. Selection rules for cutting mesh patterns:

[0035] When it is necessary to switch some SVGs from mesh mode to follow mode, the target SVGs are selected from the SVGs currently running in mesh mode according to the following priority order:

[0036] First priority: SVG whose output current is greater than the preset current percentage threshold of the rated current; the larger the current, the higher the priority.

[0037] Second priority: Among the SVGs that meet the first priority, SVGs with an electrical distance greater than the second distance threshold are selected first, and the greater the electrical distance, the higher the priority.

[0038] Third priority: When the predicted short-circuit ratio is continuously lower than the first short-circuit ratio threshold for a period of time exceeding the preset duration threshold, among the SVGs that meet the first and second priorities, the SVG with a historical cumulative network construction running time greater than the second running time threshold is selected first, and the longer the historical running time, the higher the priority.

[0039] As an optimization, the smooth transition algorithm includes three stages:

[0040] The first stage is the pre-switching preparation stage: pre-synchronization control is performed on the target SVG to make the amplitude and phase of the output voltage of the target SVG track the grid voltage, while the virtual impedance gradually changes from the current mode value to the target mode value.

[0041] The pre-switching preparation phase is initiated in advance based on the predicted short-circuit ratio: when the predicted short-circuit ratio is lower than the product of the first short-circuit ratio threshold and the preset coefficient, some network-following mode SVGs are placed in the pre-synchronization state in advance to shorten the actual handover time. The preset coefficient is a positive number less than 1.

[0042] The second stage is the mode switching stage: a hybrid coefficient is used to softly switch the control loop, while a ramp function is used to smoothly transfer the power reference value and limit the current reference value.

[0043] The third stage is the post-switching stabilization stage: monitoring voltage, current, and power fluctuations after the switch. If the fluctuations exceed a preset threshold, a rollback mechanism is triggered to restore the target SVG to the control mode before the switch.

[0044] As an optimization, the formula for calculating the virtual impedance gradient is:

[0045] ;

[0046] in, This is the virtual impedance corresponding to the current mode. The virtual impedance corresponding to the target mode. The pre-switching time is t, where t is the time variable that starts counting from the pre-switching phase.

[0047] The formula for calculating the power reference value ramp function is as follows:

[0048] ;

[0049] in, The power reference value before switching. The target power reference value, t is the transition time constant, and t is the time variable that starts counting from the mode switching phase.

[0050] As an optimization, the control parameters of the remaining SVGs are coordinated, specifically including:

[0051] The remaining SVGs operating in network-following mode are reallocated reactive power reference values ​​according to the change in the number of SVGs in network configuration mode, so that the total reactive power output of the SVG cluster meets the system's reactive power requirements:

[0052] ;

[0053] in, This is the total reactive power reference value for the SVG cluster. The number of remaining SVGs in the net pattern. The reactive power reference value for each SVG in network-connected mode. This represents the average reactive power output of a single SVG unit in a network configuration. The final required number of SVG targets for the mesh pattern;

[0054] The remaining SVGs operating in network-following mode adjust their PLL bandwidth based on the current real-time short-circuit ratio. The lower the short-circuit ratio, the lower the PLL bandwidth. The adjustment formula is as follows:

[0055] ;

[0056] in, For the adjusted PLL bandwidth, As the baseline PLL bandwidth, The current real-time short-circuit ratio, The baseline short-circuit ratio;

[0057] When the number of SVGs in network configuration mode changes, the remaining SVGs running in network configuration mode adjust their control parameters according to the new cluster configuration:

[0058] Reactive power support tasks are redistributed based on remaining capacity; SVGs with larger remaining capacity undertake more reactive power output.

[0059] ;

[0060] in, Let i be the reactive power reference value of the remaining SVG in the network. Rated capacity for a single SVG unit This represents the current used capacity of the i-th SVG.

[0061] The damping coefficient is adjusted based on the current real-time short-circuit ratio; the lower the short-circuit ratio, the larger the damping coefficient.

[0062] ;

[0063] in, Let be the damping coefficient of the remaining SVG of the i-th unit. The reference damping coefficient;

[0064] After the parameter adjustment is completed, the status registers of all network SVGs are updated synchronously to record data such as the current reactive power reference value, damping coefficient, and cumulative running time, which will be used for subsequent switching decisions.

[0065] As an optimization, the preset time constraint includes:

[0066] First time constraint: When the real-time short-circuit ratio enters the target area formed by the first short-circuit ratio threshold and the second short-circuit ratio threshold, it must continue for a first preset time length before the switching process is confirmed to be triggered.

[0067] Second time constraint: There must be a second preset time interval between two adjacent switching processes;

[0068] The value of the first preset time length is dynamically adjusted according to the predicted short-circuit ratio:

[0069] When the predicted short-circuit ratio is continuously lower than the first short-circuit ratio threshold for a period of time exceeding the first duration threshold, it is determined to be a persistent trend, and the first preset time length is shortened.

[0070] When the deviation between the predicted short-circuit ratio and the real-time short-circuit ratio exceeds a preset deviation threshold, and the predicted short-circuit ratio returns to the normal range within a second duration threshold, it is determined to be a short-term disturbance, and the first preset time length is extended or the current handover trigger is canceled.

[0071] The method further includes: when the deviation between the real-time short-circuit ratio and the predicted short-circuit ratio exceeds a preset deviation threshold, delaying the switching decision until the changing trend of the real-time short-circuit ratio is confirmed for multiple consecutive decision cycles.

[0072] This invention also discloses an SVG follow-network adaptive switching control system based on the operating short-circuit ratio, used to execute the aforementioned SVG follow-network adaptive switching control method based on the operating short-circuit ratio, comprising:

[0073] The measurement layer includes phasor measurement units deployed at grid connection points for real-time acquisition of voltage and current phasor data, and a new energy power station monitoring system for acquiring real-time power output data of each unit and predicted power output data of new energy sources.

[0074] The computing layer includes a short-circuit ratio calculation unit, a network quantity decision unit, and a coordination control unit. The short-circuit ratio calculation unit is used for multi-timescale short-circuit ratio calculation, the network quantity decision unit is used for performing SVG target quantity calculation and switching trigger decision, and the coordination control unit is used for performing target SVG selection and remaining SVG coordination control.

[0075] The execution layer consists of a cluster of multiple SVGs capable of operating in both network-building and network-following modes, with each SVG supporting smooth switching between the two control modes.

[0076] The communication layer is a high-speed communication network connecting the measurement layer, the computing layer, and the execution layer, supporting millisecond-level data transmission.

[0077] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0078] Based on the above technical solution, this invention addresses the grid intensity fluctuation problem in high-proportion renewable energy grid-connected systems by implementing adaptive switching control of multiple SVG grid-connection / grid-following modes, achieving the following technical effects (specific numerical values ​​are not involved throughout):

[0079] 1. This invention achieves real-time perception, trend capture, and forward-looking prediction of power grid strength by combining multi-timescale short-circuit ratio calculation with validity verification. It effectively filters data distortion caused by power grid transient disturbances, provides a true and continuous basis for power grid strength for subsequent decisions, avoids decision-making bias caused by single-timescale data, and improves the reliability of control logic.

[0080] 2. This invention is based on a piecewise function designed for grid strength zoning, combined with a hysteresis switching mechanism, which effectively avoids frequent switching caused by short-circuit ratio fluctuations near the threshold. At the same time, by limiting the range of the number of SVG in the grid through constraints, it ensures that the grid support meets the grid stability requirements without causing oscillations due to excessive grid construction or grid instability due to insufficient grid construction, thus achieving a balance between grid support and system stability.

[0081] 3. This invention employs a three-stage smooth transition algorithm to achieve soft switching between the control loop and the power reference value. Combined with pre-synchronization preparation and post-switching stability monitoring, it effectively suppresses voltage and current fluctuations and power surges during mode switching, ensuring that the SVG device quickly integrates into the system after switching and guaranteeing the safety and stability of the switching process.

[0082] 4. This invention enables the control logic to dynamically adapt to changes in grid strength by online adaptively adjusting key control parameters, thereby optimizing the support effect for different grid conditions. This improves the timeliness of voltage support and oscillation suppression under weak grid conditions, while ensuring the dynamic response speed under strong grid conditions, thus enhancing the adaptability and robustness of the control method.

[0083] 5. This invention selects target SVGs by prioritizing them in a hierarchical manner, taking into account both grid support effect and equipment operating status, thereby achieving a balanced lifespan for each SVG device, avoiding long-term high-load operation of some devices, and extending the overall service life of the equipment; at the same time, through coordinated control of the remaining SVG parameters, it ensures the reactive power compensation and grid support effect of the SVG cluster as a whole, and improves the operational stability and reliability of high-proportion renewable energy grid-connected systems.

[0084] 6. The entire control method of this invention does not require complex hardware modifications and can be adapted to new energy grid connection scenarios of different scales. Through flexible parameter settings and adaptive optimization, it can cope with dynamic changes in grid strength and fluctuations in new energy output, adapt to the operating characteristics of the power system, and has strong engineering practicality and promotion value. Attached Figure Description

[0085] The accompanying drawings, which are included to provide a further understanding of embodiments of the invention and form part of this application, do not constitute a limitation thereof. In the drawings:

[0086] Figure 1 This is a flowchart of an SVG tail network adaptive switching control method based on the short-circuit ratio, as described in this invention. Detailed Implementation

[0087] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the embodiments and accompanying drawings. The illustrative embodiments and descriptions of the present invention are only used to explain the present invention and are not intended to limit the present invention.

[0088] This embodiment 1 provides an adaptive switching control method for SVG trailing networks based on the short-circuit ratio, such as... Figure 1 As shown, steps S1 to S4 are included. Next, each step will be described in detail.

[0089] S1. Real-time acquisition of voltage and current phasor data and renewable energy output data at grid connection points, and calculation of the real-time short-circuit ratio of the current system based on the ratio of the inherent short-circuit capacity of the power grid to the real-time output of renewable energy.

[0090] This step constitutes the data perception and basic calculation layer of the entire control method. Its core is to achieve accurate, real-time, and reliable perception of grid strength through multi-source data acquisition, multi-timescale short-circuit ratio calculation, and validity verification. This provides a true and effective short-circuit ratio basis for subsequent grid SVG quantity decisions and switching control. The specific implementation process is divided into three sub-steps: data acquisition, multi-timescale short-circuit ratio calculation, and short-circuit ratio validity verification. The specific implementation of each sub-step is as follows:

[0091] S1.1 Real-time acquisition of multi-source data.

[0092] In this embodiment, data acquisition is accomplished through measurement layer equipment deployed in the renewable energy grid-connected system, specifically including phasor measurement units (PMUs), renewable energy power station monitoring systems (SCADA) and meteorological monitoring equipment. The acquisition content, sampling requirements, and data transmission specifications of each device are as follows:

[0093] 1. Grid Connection Point Electrical Quantity Acquisition: Phasor Measurement Units (PMUs) with a sampling rate ≥4800Hz are deployed at the new energy grid connection points to collect three-phase voltage and current phasor data in real time. The collected data is transmitted to the short-circuit ratio calculation unit at the computing layer via fiber optic Ethernet with a communication delay of ≤10ms, and is used for the equivalent impedance of the power grid. Online identification and subsequent real-time calculation of the short-circuit ratio;

[0094] 2. Real-time power output acquisition of new energy sources: Through the SCADA system of the new energy power station, the actual power output data of each photovoltaic unit and wind turbine unit in the grid-connected power station is collected in real time, and the total real-time power output of new energy sources is obtained by power aggregation. The calculation formula is: ,in For the real-time output of the i-th photovoltaic unit, To ensure the real-time output of the j-th wind turbine, the data update cycle is synchronized with the PMU to ensure the time matching between electrical quantities and output data.

[0095] 3. New Energy Output Forecasting Data Acquisition: Based on environmental data such as wind speed, light intensity, temperature, and cloud cover collected by meteorological monitoring equipment, and combined with the output forecasting model of new energy power plants, short-term output forecasting data for photovoltaic and wind power is generated. Simultaneously, the unit start-up and shutdown plans issued by the power plants are obtained through the SCADA system to correct the output forecasting results, ultimately obtaining the total new energy output forecasting data for the next 5-30 minutes. ;

[0096] 4. Obtaining the inherent short-circuit capacity of the power grid: The inherent short-circuit capacity of the power grid is calculated using the Thevenin equivalent method or an online impedance identification method based on PMU measurement data. The specific implementation is as follows:

[0097] The power grid outside the grid connection point is equivalent to the Thevenin model (equivalent impedance of an ideal voltage source in series), and the equivalent impedance is obtained through the following two methods. Method 1: Online identification based on PMU measurements. This method utilizes voltage and current phasor perturbation data acquired at high frequency by the PMU and solves the problem using the least squares method or recursive least squares method. Method 2: Based on the state estimation of the power grid dispatching system, the Thevenin equivalent impedance at the grid connection point is calculated using power flow data and topology information from the entire network. ; It can be obtained through two methods: state estimation or online identification, and can be adapted to different engineering scenarios.

[0098] Based on the obtained power grid equivalent impedance With grid connection point rated voltage The inherent short-circuit capacity of the power grid is calculated using the following formula: ;in The rated voltage of the grid at the grid connection point; the inherent short-circuit capacity of the grid is an inherent electrical parameter of the grid connection point, which remains basically stable in a short period of time. In this embodiment, the refresh and verification are performed at a 1-minute cycle, without the need for millisecond-level real-time updates.

[0099] S1.2 Calculation of short-circuit ratio at multiple time scales.

[0100] Based on the various data collected in S1.1, this embodiment adopts a multi-time-scale rolling update strategy to calculate the real-time short-circuit ratio, trend short-circuit ratio, and predicted short-circuit ratio, all of which are based on the core formula of short-circuit ratio. The derivation differs only in data source, update cycle, and calculation method, adapting to the "millisecond-level transient, second-level dynamic, and minute-level trend" change characteristics of power systems. The specific calculation rules are as follows:

[0101] Real-time short-circuit ratio calculation: The first update cycle is 100ms (which can be adjusted to 50~200ms according to actual engineering needs), based on the inherent short-circuit capacity of the power grid identified by the electrical quantities collected in real time by the PMU. And the real-time total output of new energy sources collected by the SCADA system. Calculate the real-time short-circuit ratio The calculation formula is: The real-time short-circuit ratio directly reflects the true strength of the current power grid, and the update cycle matches the control response time of power electronic equipment, ensuring millisecond-level perception of changes in power grid strength.

[0102] Trend short-circuit ratio calculation: Using 1 second as the second update period (an integer multiple of the first update period to ensure the effectiveness of data fusion), a moving average is performed on the continuously collected real-time short-circuit ratio data to calculate the trend short-circuit ratio. The formula for calculating the real-time short-circuit ratio, which filters out sudden changes in the power grid short-circuit ratio caused by transient disturbances (such as lightning strikes, sudden increases or decreases in load, and short-term start-ups and shutdowns of generating units), reflects the continuous trend of power grid strength. ;in, The moving average coefficient is set to 0.2 in this embodiment (it can be adjusted between 0.1 and 0.3 according to engineering requirements). This is the calculated value of the trend short-circuit ratio for the previous period; the moving average processing can effectively smooth the high-frequency fluctuations of the real-time short-circuit ratio and preserve the true trend of power grid strength changes.

[0103] Predicted short-circuit ratio calculation: with a third update cycle of 5 minutes, based on the refreshed inherent short-circuit capacity of the power grid. And new energy power output forecast data Calculate the predicted short-circuit ratio This method predicts the changing trend of power grid strength over a future period, providing a forward-looking basis for adjusting the time constraints of subsequent handover decisions and preparing for pre-handover. The calculation formula is as follows: The update cycle for predicting the short-circuit ratio is matched with the time scale of the short-term output prediction of new energy sources, taking into account both prediction accuracy and foresight, and avoiding prediction failure due to too short a prediction cycle and excessive prediction error due to too long a prediction cycle.

[0104] S1.3 Real-time short-circuit ratio validity verification.

[0105] Because the real-time short-circuit ratio has a short update cycle and is sensitive to transient disturbances, it is prone to data distortion due to short-term anomalies in the power grid. Directly using distorted data as the basis for subsequent decisions can lead to errors in the calculation of the number of SVG (Static Var Generator) in the network and misjudgments in switching decisions. Therefore, this embodiment uses the trend short-circuit ratio as a verification benchmark to verify the validity of the real-time short-circuit ratio, eliminating invalid data and ensuring the authenticity and reliability of the subsequent decision-making basis. The specific verification rules are as follows:

[0106] Set preset deviation threshold In this embodiment, 0.3 is used (which can be adjusted between 0.2 and 0.5 according to the characteristics of the grid at the grid connection point) as the maximum allowable deviation between the real-time short-circuit ratio and the trend short-circuit ratio;

[0107] The real-time short-circuit ratio is calculated each time. Then, immediately calculate its short-circuit ratio with the current trend. The absolute difference, i.e. ;like The real-time short-circuit ratio calculated in this study is deemed valid data and will be used as the baseline value for the current system's short-circuit ratio for subsequent quantitative calculations of the number of SVG units in the network. If the real-time short-circuit ratio calculated in this instance is deemed invalid (due to transient disturbance), the result of this calculation is discarded, and the previous valid real-time short-circuit ratio value is maintained until the difference between the real-time short-circuit ratio and the trend short-circuit ratio returns to within the preset deviation threshold.

[0108] The validity verification is performed simultaneously with the real-time short-circuit ratio calculation. The verification result is transmitted to the network quantity decision unit of the computing layer along with the real-time short-circuit ratio, ensuring that the decision unit always obtains the verified valid short-circuit ratio data.

[0109] In this embodiment, by combining multi-timescale short-circuit ratio calculation with validity verification, real-time perception of grid strength changes is ensured, transient disturbances are filtered out by trend short-circuit ratio, and forward-looking prediction is achieved by predicting the short-circuit ratio. This provides accurate, reliable, and complete short-circuit ratio data support for quantitative decision-making and adaptive switching in subsequent steps, effectively avoiding decision-making errors caused by single-timescale data.

[0110] S2. Based on the relationship between the real-time short-circuit ratio and the first short-circuit ratio threshold and the second short-circuit ratio threshold, under the constraints of the minimum network ratio and the maximum network ratio, the number of SVG targets operating in the required network mode is calculated using a piecewise function, wherein the first short-circuit ratio threshold is less than the second short-circuit ratio threshold.

[0111] This step is the core decision-making layer of the entire control method. Based on the effective real-time short-circuit ratio obtained from S1, combined with the preset short-circuit ratio threshold and the minimum / maximum network ratio constraint, a piecewise function is used to quantitatively calculate the target number of SVGs required to operate in the network configuration mode. This provides a quantitative basis for subsequent switching triggering and mode adjustment. The specific implementation process is divided into three sub-steps: threshold setting, piecewise function calculation, and constraint verification.

[0112] S2.1 Key Thresholds and Parameter Settings.

[0113] This embodiment pre-sets the following core thresholds and parameters to adapt to the grid characteristics of scenarios with a high proportion of renewable energy connected to the grid:

[0114] 1. Short-circuit ratio threshold:

[0115] First short-circuit ratio threshold (Critical value for weak power grids; emergency support is required if the value falls below this).

[0116] Second short-circuit ratio threshold (Critical value for a strong power grid; above this value, only minimum support is required.)

[0117] Target short-circuit ratio (Optimal short-circuit ratio for system operation);

[0118] 2. Network ratio constraints:

[0119] Minimum network ratio (Ensure basic voltage support and avoid operation without a grid).

[0120] Maximum network ratio (Avoid excessive network construction that could cause high-frequency oscillations);

[0121] 3. SVG basic parameters:

[0122] Total number of SVG units ;

[0123] Rated capacity of a single SVG unit ;

[0124] Network support coefficient (Reflects the equivalent short-circuit capacity contribution of a single SVG unit in a network).

[0125] 4. Quantity difference threshold: Preset quantity difference threshold (In this embodiment, we take 1), which is used to determine whether the switching process is triggered.

[0126] 5. Hysteresis and Adaptive Related Parameters: Hysteresis Width Initial value 0.2 (adjustable), reference transition time constant Initial value 50ms (adjustable), reference damping coefficient Baseline PLL bandwidth Hz, reference short-circuit ratio .

[0127] S2.2 Definition of key constraints.

[0128] Before quantitatively calculating the target number of SVG meshes, this embodiment clarifies the following three types of key constraints to ensure that the number of meshes is within a reasonable range and to avoid extreme operating conditions:

[0129] Upper limit constraint: This is used to limit the maximum number of SVGs used in mesh construction, preventing excessive mesh construction from causing high-frequency oscillations. Typically, the value is 0.6 to 0.8; in this embodiment, it is 0.7.

[0130] Lower bound constraint: Used to ensure basic voltage support for the power grid and prevent grid instability caused by unstructured grid operation. Typically, the value is taken as 0.2 to 0.3; in this embodiment, it is taken as 0.2.

[0131] Integer constraints: The number of SVGs in the network must be an integer, not exceeding the total number of units and not less than 0, in accordance with the actual equipment switching logic of the project.

[0132] S2.3 Derivation of total system short-circuit capacity and target short-circuit ratio.

[0133] In this embodiment, the total short-circuit capacity of the system is composed of two parts connected in parallel: the inherent short-circuit capacity of the power grid and the equivalent short-circuit capacity of the grid SVG. This provides a physical basis for the quantitative calculation of the number of grid SVGs.

[0134] 1. Inherent short-circuit capacity of the power grid.

[0135] Thevenin equivalent method or an online impedance identification method based on PMU measurement is used, through the rated voltage at the grid connection point. Equivalent impedance to power grid calculate:

[0136] ;in, The magnitude of the equivalent impedance of the power grid can be obtained through state estimation or online identification. It reflects the inherent strength of the power grid at the grid connection point and remains basically stable in a short period of time. In this embodiment, the verification is refreshed every 1 minute.

[0137] 2. Equivalent short-circuit capacity of SVG network.

[0138] An SVG operating in network mode can be considered an equivalent voltage source, providing additional short-circuit capacity support for the system. Its equivalent short-circuit capacity is calculated using the following formula: ;in, The number of SVGs currently running in network configuration mode; The rated capacity of a single SVG unit; The network support coefficient reflects the enhancement effect of network control on the system's short-circuit capacity; in this embodiment, it is set to 1.3.

[0139] 3. Total short-circuit capacity of the system.

[0140] The total short-circuit capacity of the system is the sum of the inherent short-circuit capacity of the power grid and the equivalent short-circuit capacity of the grid-connected SVG: .

[0141] 4. Target short-circuit ratio Derivation of the theory of network quantity.

[0142] Target short-circuit ratio Defined as the total short-circuit capacity of the system and the real-time output of new energy sources. The ratio is used to ensure stable system operation: Based on the inverse solution of this formula, the theoretical formula for the number of SVG network components is obtained after simplification: This formula provides the core physical basis for subsequent quantitative calculations of piecewise functions.

[0143] S2.4 Calculate the number of SVG targets in the network using piecewise functions.

[0144] To avoid frequent switching due to small fluctuations in the short-circuit ratio around the threshold, this embodiment introduces a hysteresis switching mechanism to effectively monitor the short-circuit ratio in real time. As a criterion for judgment, combined with hysteresis width A buffered switching decision logic is formed, and then the target number of network SVG is calculated using a piecewise function according to the power grid state partition. The details are as follows:

[0145] 1. Hysteresis switching determination.

[0146] Set hysteresis width (Initial value 0.2, can be adjusted according to project requirements), serving as a buffer for the short-circuit ratio threshold: when the effective real-time short-circuit ratio Crossing the second short-circuit ratio threshold from low to high At that time, it is necessary to meet the following requirements. Only then does it proceed to the subsequent quantity difference calculation and switching triggering process; when the effective real-time short-circuit ratio Crossing the first short-circuit ratio threshold from high to low At that time, it is necessary to meet the following requirements. Only then does it proceed to the subsequent quantity difference calculation and switching triggering process; within the hysteresis zone ( The current number of SVGs in the network remains unchanged, and no recalculation or switching operations are triggered.

[0147] 2. Calculate the target quantity using piecewise functions.

[0148] After the hysteresis limit condition is met, the inherent short-circuit capacity of the power grid is considered. Real-time power output with new energy sources Based on the relationship, the power grid state is divided into three regions, and the target number of network SVG is calculated using a piecewise function. :

[0149] Scenario 1: Emergency Support Area ;

[0150] At this point, the power grid strength is extremely weak, and even with a full SVG (Static Var Generator) grid, the target short-circuit ratio cannot be achieved; therefore, the maximum permissible grid capacity must be utilized.

[0151] ;

[0152] Scenario 2: Quantitative Calculation Area ;

[0153] At this point, the grid strength is moderate. First, calculate the theoretical number of SVG (Static Var Generator) for the grid using theoretical formulas, then apply upper limit constraints, lower limit constraints, and integer constraints, and round up:

[0154] Theoretical quantitative formula:

[0155] ;

[0156] Number of targets after constraints:

[0157] ;

[0158] in The rounding symbol is used to ensure that the number of nets is an integer.

[0159] Case 3: Minimum Support Area ;

[0160] At this point, the power grid strength is sufficient, and only the minimum grid connection ratio needs to be maintained to ensure basic support:

[0161] ;

[0162] S2.5 Adaptive parameter optimization.

[0163] During stable system operation, this embodiment uses reinforcement learning or fuzzy logic methods to adaptively optimize key control parameters online based on historical operating data. The optimization cycle is 5 minutes per cycle, and the optimized parameters are synchronously updated to all SVG status registers, providing an optimization basis for subsequent decision-making and control. The specific optimization rules are as follows:

[0164] 1. Hysteresis width Adaptive adjustment:

[0165] If the overshoot of the grid connection point voltage is greater than 5% after switching, it indicates that the current hysteresis width is too small and the switching is too sensitive, and it needs to be increased. This further suppresses frequent switching and voltage fluctuations;

[0166] If the system response is slow and voltage support is insufficient after the switch, it indicates that the current hysteresis width is too large and needs to be appropriately reduced. This ensures a timely response to weak power grids.

[0167] 2. Adaptive adjustment of short-circuit ratio threshold and network configuration ratio:

[0168] If low SCR ( If the duration is too long, it indicates that the current threshold or network configuration ratio is too conservative and needs to be appropriately reduced. Pre-trigger SVG deployment for network construction; or increase This allows for the deployment of more SVG (Static Var Generator) in the network, strengthening the support capabilities of weak power grids.

[0169] 3. Transition time constant Adaptive adjustment:

[0170] If a power / voltage oscillation event occurs in the system, it indicates that the current mode switching transition process is too fast or too slow, and the transition time constant needs to be adjusted. When the oscillation is violent, increase Extend the smooth transition time between the pre-switching and mode switching phases; reduce the response time when the response is slow. This accelerates the adjustment speed of control modes and parameters.

[0171] S2.6 Comparison of target quantity and current quantity.

[0172] Calculate the target quantity Then, the number of SVGs currently actually running in network configuration mode. Comparison:

[0173] ;

[0174] like : Determine that no switching is needed, maintain the current network configuration / follow-the-network mode distribution;

[0175] like : Proceed to the subsequent switching trigger and target SVG selection process, The value is usually 1.

[0176] S3. Compare the number of SVGs currently running in network construction mode with the target number. When the absolute value of the difference between the number of SVGs currently running in network construction mode and the target number exceeds a preset number threshold and satisfies a preset time constraint, determine whether to switch some SVGs in network following mode to network construction mode or switch some SVGs in network construction mode to network following mode based on the sign of the difference. Select the target SVG according to the preset rules and trigger the switching process.

[0177] This step is the action triggering layer of the control method. Based on the quantity difference obtained from S2, it determines whether to trigger a switch based on dynamic time constraints, and selects the target SVG according to priority rules to ensure the rationality of the switch and the balance between equipment lifespan. The specific implementation process is divided into two sub-steps: switch triggering decision and target SVG selection.

[0178] S3.1 Dynamic Time Constraints and Switching Triggers.

[0179] To avoid frequent handovers caused by short-term disturbances, this embodiment is based on the core idea of ​​validity verification + auxiliary judgment + adaptive parameter optimization, and uses dynamic time constraints to determine whether to trigger a handover. The specific rules are as follows:

[0180] (1) First-time constraints: trend confirmation and disturbance filtering.

[0181] When the effective real-time short-circuit ratio enters the first short-circuit ratio threshold Compared with the second short-circuit threshold After the target area is defined, it needs to be continuously maintained for a first preset time length T1 (usually 3 to 5 decision cycles) before the switching process is confirmed to filter transient disturbances and confirm the trend of power grid strength changes.

[0182] The value of T1 is not fixed, but is adaptively adjusted based on the changing trend of the predicted short-circuit ratio: if the predicted short-circuit ratio is continuously lower than... Exceeding the first duration threshold If the short circuit ratio is continuously low for too long, it is determined to be a persistent weak grid trend. Based on the adaptive optimization idea of ​​strengthening support when the low SCR is continuously low for too long, T1 is shortened (e.g., from 3s to 1s) to accelerate the response speed of the grid SVG. If the deviation between the predicted short circuit ratio and the effective real-time short circuit ratio exceeds the preset deviation threshold, and the predicted short circuit ratio returns to the normal range within the second time threshold, it is determined to be a short-term disturbance. Referring to the validity verification logic of removing outliers and continuous confirmation, T1 is extended (e.g., from 3s to 5s) or the current handover trigger is directly canceled to avoid invalid actions.

[0183] (2) Second time constraint: transient process avoidance.

[0184] A second preset time length T2 (30s in this embodiment, which can be adjusted between 30 and 60s according to engineering requirements) is required between two adjacent switching processes to avoid the system transient process after switching and prevent frequent switching from causing instability of the power grid or SVG equipment.

[0185] (3) Delayed decision-making mechanism when the deviation is too large.

[0186] If the deviation between the effective real-time short-circuit ratio and the predicted short-circuit ratio continues to exceed the preset deviation threshold, and the trend of power grid strength change cannot be clearly determined, the switching decision is delayed in accordance with the abnormal value handling logic of validity verification + multi-cycle confirmation until the real trend of the effective real-time short-circuit ratio is confirmed in multiple consecutive decision cycles (such as 5 cycles) before the switching decision is made, so as to avoid incorrect decision-making due to prediction deviation or transient anomaly.

[0187] Satisfying the above time constraints and At that time, the switching process is officially triggered: if : need to The SVG in platform and network mode switches to network construction mode; if : need to The SVG in the platform network mode is switched to the network mode.

[0188] S3.2 Target SVG Priority Selection.

[0189] Depending on the switching direction, a hierarchical priority rule is used to select the target SVG, balancing grid support effectiveness with equipment lifespan.

[0190] (1) Following the network cutting structure ( Selection rules.

[0191] When it is necessary to increase the number of network SVGs, select them from the current network mode SVGs in descending order of priority as follows. tower:

[0192] Device status priority: Prioritize SVGs that are currently in network mode, in good health (no fault alarms, no overheating, no communication abnormalities), and have remaining capacity > 30% threshold to ensure safe and reliable device deployment;

[0193] Electrical distance priority: Among the SVGs that meet the first priority, the SVGs that are closer to the grid connection point are selected first to reduce the impact of line voltage drop and improve the support effect on the power grid;

[0194] Historical operation priority: If the predicted short-circuit ratio is consistently weak for more than the preset duration, the SVG with the shorter historical cumulative network operation time will be selected first to balance the lifespan of each SVG device and avoid some devices from operating under high load for a long time.

[0195] (2) Network construction and cutting ( Selection rules.

[0196] When it is necessary to reduce the number of meshing SVGs, select from the current meshing mode SVGs in descending order of priority as follows. tower:

[0197] Operating load priority: Prioritize SVGs that are currently in grid mode and whose output current is close to the limit (>80% of rated current) to avoid long-term overload operation of the equipment;

[0198] Electrical distance priority: Among the SVGs that meet the first priority, the SVGs with a greater electrical distance from the grid connection point are selected first to reduce the transient impact of the disconnection operation on the core power grid;

[0199] Historical operation priority: If the predicted short circuit ratio is consistently stronger than the preset duration, the SVG with the longer historical cumulative network operation time will be selected first, allowing the equipment that has been operating in the network for a long time to take turns to rest and balance the overall lifespan.

[0200] S4. Using a smooth transition algorithm, switch the target SVG from the current control mode to the target control mode, and coordinate the control parameters of the remaining SVGs.

[0201] This step is the execution and optimization layer of the control method. It uses a three-stage smooth transition algorithm to achieve shock-free mode switching of the target SVG and simultaneously adjusts the control parameters of the remaining SVGs to ensure the overall compensation effect of the SVG cluster. The specific implementation process is divided into two sub-steps: smooth mode switching and coordinated control of the remaining SVGs.

[0202] S4.1, Three-stage smooth mode switching.

[0203] For the selected target SVG, a three-stage smooth transition algorithm of pre-switching preparation - mode switching - post-switching stabilization is adopted to avoid current surges and voltage fluctuations during the switching process.

[0204] Phase 1: Pre-switch preparation ( , ms).

[0205] The target SVG enters pre-synchronization state: the output voltage amplitude and phase accurately track the grid voltage;

[0206] Virtual impedance linear gradient: from current mode value To target mode value Gradient, the formula is:

[0207] ;

[0208] in The pre-switching time is t, where t is the timing variable within the phase.

[0209] If the short-circuit ratio is predicted Below If the grid is about to enter an emergency support state, the pre-synchronization process is initiated in advance, and the pre-switching preparation stage is moved forward. This allows the target SVG to complete voltage tracking and impedance gradation before the formal switching command is issued, reducing the actual switching response time from 500ms to less than 100ms and significantly improving the timeliness of voltage support under weak grid conditions.

[0210] Phase 2: Mode Switching ).

[0211] Control loop soft switching: A hybrid coefficient smooth transition strategy is adopted to achieve bidirectional soft switching between the Virtual Synchronous Machine (VSM) loop and the Phase-Locked Loop (PLL) loop, avoiding power surges caused by sudden changes in control mode: When switching from grid to network construction: the weight of the PLL loop is gradually reduced and the weight of the VSM loop is increased, so that the SVG smoothly transitions from current source control to voltage source control; When switching from network construction to grid connection: the weight of the VSM loop is gradually reduced and the weight of the PLL loop is increased, so that the SVG smoothly transitions from voltage source control to current source control; the hybrid coefficient changes linearly with time, and the formula is: , ,in For the pre-switching time, This is the mode switching time, i.e., the transition time constant (initially set to 50ms, which can be adaptively adjusted). The control loop is linearly increased from 0 to 1 to ensure smooth switching without any impact.

[0212] Power reference value ramp transition: In conjunction with control loop switching, the power reference value is smoothed by a ramp to avoid sudden power changes. The formula is as follows: ;in , These are the power reference values ​​before and after the switching;

[0213] Current limiting protection: Limits the current reference value to within 1.2 times the rated current to prevent surges. , This is the current reference value. This is the rated current value.

[0214] Phase 3: Post-switch stabilization ( back).

[0215] The system monitors the SVG output voltage, current, power, and grid connection point short-circuit ratio in real time to determine whether the system has entered a stable operating state. When the SVG output fluctuation is less than the 3% threshold and the grid connection point voltage deviation is less than the 2% threshold for more than 100ms, the system is determined to have entered a stable state. The SVG status register is then updated, and key statuses such as the target SVG's current operating mode (networking / following the network) and cumulative networking runtime are written into the register to provide the latest status basis for subsequent decisions and switching. If abnormalities such as voltage / current fluctuations exceeding limits or the short-circuit ratio not returning to the target range are found during the stability monitoring period, the rollback mechanism is immediately triggered to restore the target SVG to the mode before the switchover, ensuring system safety.

[0216] S4.2 Residual SVG Coordination Control and Parameter Adaptive Adjustment.

[0217] After the mode switch is completed, the parameters of the remaining SVGs running in network construction mode and network following mode are coordinated to ensure the overall reactive power compensation of the cluster, the grid support effect, and the stability of weak grids. The specific implementation is as follows:

[0218] (1) Remaining SVG parameter coordination:

[0219] After the mode switch is completed, the reactive power reference value is reallocated and the phase-locked loop (PLL) bandwidth is dynamically adjusted to adapt to changes in grid strength.

[0220] 1. Reactive power reference value redistribution.

[0221] Based on the changes in the number of SVGs in the network, the reactive power reference values ​​of the network SVGs are reallocated to ensure that the total reactive power of the cluster meets the system requirements. ;in: This is the total reactive power reference value for the SVG cluster; This represents the remaining number of SVGs in the web interface; This represents the average reactive power output of a single SVG unit in a network configuration.

[0222] 2. PLL bandwidth is dynamically adjusted.

[0223] Based on the current effective real-time short-circuit ratio, the bandwidth of the SVG phase-locked loop is adjusted. In weak grid conditions, the bandwidth is reduced to avoid instability. Therefore, the adjusted PLL bandwidth is: ;in: The baseline PLL bandwidth (100Hz in this embodiment); The short-circuit ratio is taken as the baseline (2.0 in this embodiment); the lower the short-circuit ratio, the smaller the PLL bandwidth, which improves the stability under weak power grid conditions.

[0224] (2) Adaptive adjustment of remaining mesh SVG parameters

[0225] When the number of SVGs in the network configuration changes, the remaining SVGs need to be reallocated for reactive power support tasks and the damping coefficient dynamically adjusted according to the new cluster configuration:

[0226] 1. Reactive power support task reallocation:

[0227] Based on the remaining capacity of each network SVG, the total reactive power support task of the cluster is redistributed, with SVGs with larger remaining capacity undertaking more reactive power output:

[0228] ;in: Let i be the reactive power reference value of the remaining SVG in the network. Rated capacity for a single SVG unit This represents the current used capacity of the i-th SVG. This represents the sum of the remaining capacity of all remaining SVG structures.

[0229] 2. Adaptive adjustment of damping coefficient:

[0230] Based on the current effective real-time short-circuit ratio Adjusting the damping coefficient of the network SVG virtual synchronizing machine (VSM) increases the damping coefficient as the short-circuit ratio decreases, thereby enhancing the oscillation suppression capability under weak grid conditions. ;in, The damping coefficient of the remaining SVG of the i-th unit is adjusted. The reference damping coefficient is 0.8 in this embodiment. The baseline short-circuit ratio is 2.0 (in this embodiment, it is taken as 2.0); when hour, Increase damping to suppress oscillations; when hour, This ensures dynamic response speed.

[0231] (3) Status register synchronous update:

[0232] After completing parameter adjustments for all remaining SVGs (mesh + follow mesh), synchronously update the status registers of all SVGs:

[0233] For network SVG: Record the current reactive power reference value Damping coefficient Cumulative network runtime;

[0234] For the SVG (Static Var Generator): Records the current reactive power reference value. PLL bandwidth ;

[0235] This provides the latest status information for subsequent switching decisions and control logic.

[0236] For ease of understanding, recommended values ​​for key parameters are shown in Table 1:

[0237] Table 1

[0238]

[0239] The method of the present invention will be described next through specific examples.

[0240] First, the SCR status is partitioned, as shown in Table 2:

[0241] Table 2

[0242]

[0243] Example 1: SVG Cluster Control of Photovoltaic Base

[0244] Scene description:

[0245] Installed capacity: 500MW photovoltaic + 100Mvar SVG cluster (10 units × 10Mvar);

[0246] Power grid connection: 220kV transmission line, equivalent impedance pu (weak grid scenario);

[0247] Parameter settings:

[0248] Calculation process:

[0249] Inherent short-circuit capacity of the power grid: ;

[0250] Time 1: 12:00 noon, photovoltaic output PPV = 450 MW (90%);

[0251] (Area A: Emergency Support);

[0252] Decision: Deploy maximum network capacity (10 × 0.7);

[0253] verify: ; Although it is still below 1.5, it has provided maximum support and needs to be combined with other measures.

[0254] Time 2: 10:00 AM, photovoltaic output PPV = 250 MW (50%); (Area B: Quantitative Calculation);

[0255] Calculate the number of SVGs in the mesh: ;

[0256] Constraint handling: (Take the minimum value);

[0257] Decision: Retain 2 SVG units for network construction (minimum support), and the remaining 8 units will be connected to the network.

[0258] Time 3: 18:00 in the evening, photovoltaic output PPV = 50 MW (10%);

[0259] (Category C: Minimum Support);

[0260] Decision: Retain the minimum network configuration ratio .

[0261] Using the method of this invention, the SVG cluster adjusts the network construction / follow-the-network ratio in real time according to SCR; during the high-output period at noon, 7 SVGs are in network construction operation to provide equivalent inertia support, and the voltage fluctuation is <5%; during the low-output period in the evening, only 2 SVGs are in network construction operation to avoid overdamped oscillation; the switching process is smooth, the transient voltage deviation is <3%, and the recovery time is <200ms.

[0262] Example 2: Suppression of broadband oscillations in offshore wind farms

[0263] Scene description:

[0264] Installed capacity: 800MW offshore wind power;

[0265] Grid connection method: Flexible DC transmission (VSC-HVDC), with an equivalent SCR variation range of 1.2~4.0;

[0266] SVG configuration: 8 units × 20Mvar , .

[0267] Using this invention, the AC-side SCR of the converter station is monitored in real time; when SCR < 1.5, 6 grid SVG units (maximum allowable) are put into operation to provide positive damping; when SCR > 3.0, 2 grid SVG units (minimum support) are retained to avoid oscillation; the damping controller parameters are adaptively adjusted according to SCR.

[0268] Comparison of effects:

[0269] Fixed tracking mode: Subsynchronous oscillation occurs when SCR=1.3, and the amplitude decay time is >10s;

[0270] Fixed network configuration: High-frequency oscillations occur at SCR=3.5, and voltage THD>8%;

[0271] The adaptive mode of this invention: stable operation throughout, voltage deviation <2%, and no oscillation phenomenon.

[0272] Example 2 also discloses an SVG follow-network adaptive switching control system based on the operating short-circuit ratio, used to execute the SVG follow-network adaptive switching control method based on the operating short-circuit ratio described in Example 1, including:

[0273] The measurement layer includes phasor measurement units deployed at grid connection points for real-time acquisition of voltage and current phasor data, and a new energy power station monitoring system for acquiring real-time power output data of each unit and predicted power output data of new energy sources.

[0274] The computing layer includes a short-circuit ratio calculation unit, a network quantity decision unit, and a coordination control unit. The short-circuit ratio calculation unit is used for multi-timescale short-circuit ratio calculation, the network quantity decision unit is used for performing SVG target quantity calculation and switching trigger decision, and the coordination control unit is used for performing target SVG selection and remaining SVG coordination control.

[0275] The execution layer consists of a cluster of multiple SVGs capable of operating in both network-building and network-following modes, with each SVG supporting smooth switching between the two control modes.

[0276] The communication layer is a high-speed communication network connecting the measurement layer, the computing layer, and the execution layer, supporting millisecond-level data transmission.

[0277] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. An adaptive switching control method for SVG trailing networks based on short-circuit ratio, characterized in that, include: Real-time collection of voltage and current phasor data and renewable energy output data at grid connection points; calculation of the real-time short-circuit ratio of the current system based on the ratio of the grid's inherent short-circuit capacity to the real-time output of renewable energy. Based on the relationship between the real-time short-circuit ratio and the first short-circuit ratio threshold and the second short-circuit ratio threshold, under the constraints of the minimum network ratio and the maximum network ratio, the number of SVG targets running in the required network mode is calculated using a piecewise function, where the first short-circuit ratio threshold is less than the second short-circuit ratio threshold. The calculation of the number of SVG targets for the required network configuration using a piecewise function specifically includes: When the inherent short-circuit capacity of the power grid is less than the product of the first short-circuit ratio threshold and the real-time output of new energy, the product of the maximum grid construction ratio and the total number of SVG units is taken as the target number. When the inherent short-circuit capacity of the power grid is greater than the product of the second short-circuit ratio threshold and the real-time output of new energy, the product of the minimum grid construction ratio and the total number of SVG units is taken as the target number. When the inherent short-circuit capacity of the power grid is between the product of the first short-circuit ratio threshold and the real-time output of new energy sources and the product of the second short-circuit ratio threshold and the real-time output of new energy sources, the target quantity is calculated using a theoretical formula, and then rounded after applying the constraints of the minimum and maximum grid construction ratios. The first short-circuit ratio threshold is the critical value for a weak power grid, and the second short-circuit ratio threshold is the critical value for a strong power grid; Compare the number of SVGs currently running in network construction mode with the target number. When the absolute value of the difference between the number of SVGs currently running in network construction mode and the target number exceeds a preset number threshold and meets a preset time constraint, determine whether to switch some network-following mode SVGs to network construction mode or switch some network-following mode SVGs to network-following mode based on the sign of the difference. Select the target SVG according to the preset rules and trigger the switching process. The target SVG is switched from the current control mode to the target control mode using a smooth transition algorithm, and the control parameters of the remaining SVGs are coordinated.

2. The adaptive switching control method for SVG trailing network based on short-circuit ratio according to claim 1, characterized in that, The calculation of the real-time short-circuit ratio includes multiple time scales: The real-time short-circuit ratio is calculated based on the data collected in real time by the phasor measurement unit, with the first update cycle. The real-time short-circuit ratio is averaged, and the trend short-circuit ratio is calculated with the second update period; Based on the new energy power output forecast data, the predicted short-circuit ratio is calculated in the third update cycle. The new energy power output forecast data includes at least photovoltaic power output forecast and wind power output forecast based on meteorological data. The validity of the real-time short-circuit ratio is verified. When the difference between the real-time short-circuit ratio and the trend short-circuit ratio exceeds a preset deviation threshold, it is determined to be invalid data, and the previous valid value is maintained.

3. The adaptive switching control method for SVG trailing network based on short-circuit ratio according to claim 1, characterized in that, When the inherent short-circuit capacity of the power grid is between the product of the first short-circuit ratio threshold and the real-time output of new energy sources, and the second short-circuit ratio threshold and the real-time output of new energy sources, the target quantity is calculated using a theoretical formula: ； in, For the target short-circuit ratio, Rated capacity for a single SVG unit. For the network support coefficient, For the inherent short-circuit capacity of the power grid, Providing real-time power for new energy sources The theoretical computational cost of the SVG in the network configuration mode; The target number after applying minimum and maximum meshing ratio constraints and rounding is: ； in, To achieve the maximum network ratio, To minimize the network ratio, This represents the total number of SVG units. The final number of SVG targets for the desired mesh pattern.

4. The SVG trailing network adaptive switching control method based on the short-circuit ratio according to claim 1, characterized in that, The target SVG is selected according to preset rules, specifically including: A. Selection rules when deploying network configuration modes: When it is necessary to switch some SVGs from mesh mode to mesh construction mode, the target SVGs are selected from the SVGs currently running in mesh mode according to the following priority order: First priority: SVG with no fault alarms, no overheating, no communication abnormalities, and remaining capacity greater than the remaining capacity threshold; Second priority: Among the SVGs that meet the first priority, SVGs with an electrical distance less than the first distance threshold are selected first, and the smaller the electrical distance, the higher the priority. Third priority: When the predicted short-circuit ratio is continuously lower than the first short-circuit ratio threshold for a period of time exceeding the preset duration threshold, among the SVGs that meet the first and second priorities, the SVGs whose historical cumulative network construction running time is less than the first running time threshold are selected first, and the shorter the historical running time, the higher the priority. B. Selection rules for cutting mesh patterns: When it is necessary to switch some SVGs from mesh mode to follow mode, the target SVGs are selected from the SVGs currently running in mesh mode according to the following priority order: First priority: SVG whose output current is greater than the preset current percentage threshold of the rated current; the larger the current, the higher the priority. Second priority: Among the SVGs that meet the first priority, SVGs with an electrical distance greater than the second distance threshold are selected first, and the greater the electrical distance, the higher the priority. Third priority: When the predicted short-circuit ratio is continuously lower than the first short-circuit ratio threshold for a period of time exceeding the preset duration threshold, among the SVGs that meet the first and second priorities, the SVG with a historical cumulative network construction running time greater than the second running time threshold is selected first, and the longer the historical running time, the higher the priority.

5. The SVG trailing network adaptive switching control method based on the short-circuit ratio according to claim 2, characterized in that, The smooth transition algorithm consists of three stages: The first stage is the pre-switching preparation stage: pre-synchronization control is performed on the target SVG to make the amplitude and phase of the output voltage of the target SVG track the grid voltage, while the virtual impedance gradually changes from the current mode value to the target mode value. The pre-switching preparation phase is initiated in advance based on the predicted short-circuit ratio: when the predicted short-circuit ratio is lower than the product of the first short-circuit ratio threshold and the preset coefficient, some network-following mode SVGs are placed in the pre-synchronization state in advance to shorten the actual handover time. The preset coefficient is a positive number less than 1. The second stage is the mode switching stage: a hybrid coefficient is used to softly switch the control loop, while a ramp function is used to smoothly transfer the power reference value and limit the current reference value. The third stage is the post-switching stabilization stage: monitoring voltage, current, and power fluctuations after the switch. If the fluctuations exceed a preset threshold, a rollback mechanism is triggered to restore the target SVG to the control mode before the switch.

6. The SVG trailing network adaptive switching control method based on the short-circuit ratio according to claim 5, characterized in that, The formula for calculating the virtual impedance gradient is: ； in, This is the virtual impedance corresponding to the current mode. The virtual impedance corresponding to the target mode. The pre-switching time is t, where t is the time variable that starts counting from the pre-switching phase. The formula for calculating the power reference value ramp function is as follows: ； in, The power reference value before switching. The target power reference value, t is the transition time constant, and t is the time variable that starts counting from the mode switching phase.

7. The adaptive switching control method for SVG trailing network based on short-circuit ratio according to claim 1, characterized in that, Coordinate the control parameters of the remaining SVGs, specifically including: The remaining SVGs operating in network-following mode are reallocated reactive power reference values ​​according to the change in the number of SVGs in network configuration mode, so that the total reactive power output of the SVG cluster meets the system's reactive power requirements: ； in, This is the total reactive power reference value for the SVG cluster. The number of remaining SVGs in the net pattern. The reactive power reference value for each SVG in network-connected mode. This represents the average reactive power output of a single SVG unit in a network configuration. The final required number of SVG targets for the mesh pattern; The remaining SVGs operating in network-following mode adjust their PLL bandwidth based on the current real-time short-circuit ratio. The lower the short-circuit ratio, the lower the PLL bandwidth. The adjustment formula is as follows: ； in, For the adjusted PLL bandwidth, As the baseline PLL bandwidth, The current real-time short-circuit ratio, The baseline short-circuit ratio; When the number of SVGs in network configuration mode changes, the remaining SVGs running in network configuration mode adjust their control parameters according to the new cluster configuration: Reactive power support tasks are redistributed based on remaining capacity; SVGs with larger remaining capacity undertake more reactive power output. ； in, Let i be the reactive power reference value of the remaining SVG in the network. Rated capacity for a single SVG unit This represents the current used capacity of the i-th SVG. The damping coefficient is adjusted based on the current real-time short-circuit ratio; the lower the short-circuit ratio, the larger the damping coefficient. ； in, Let be the damping coefficient of the remaining SVG of the i-th unit. The reference damping coefficient; After the parameter adjustment is completed, the status registers of all network SVGs are updated synchronously to record data such as the current reactive power reference value, damping coefficient, and cumulative running time, which will be used for subsequent switching decisions.

8. The SVG trailing network adaptive switching control method based on the short-circuit ratio according to claim 2, characterized in that, The preset time constraints include: First time constraint: When the real-time short-circuit ratio enters the target area formed by the first short-circuit ratio threshold and the second short-circuit ratio threshold, it must continue for a first preset time length before the switching process is confirmed to be triggered. Second time constraint: There must be a second preset time interval between two adjacent switching processes; The value of the first preset time length is dynamically adjusted according to the predicted short-circuit ratio: When the predicted short-circuit ratio is continuously lower than the first short-circuit ratio threshold for a period of time exceeding the first duration threshold, it is determined to be a persistent trend, and the first preset time length is shortened. When the deviation between the predicted short-circuit ratio and the real-time short-circuit ratio exceeds a preset deviation threshold, and the predicted short-circuit ratio returns to the normal range within a second duration threshold, it is determined to be a short-term disturbance, and the first preset time length is extended or the current handover trigger is canceled. The method further includes: when the deviation between the real-time short-circuit ratio and the predicted short-circuit ratio exceeds a preset deviation threshold, delaying the switching decision until the changing trend of the real-time short-circuit ratio is confirmed for multiple consecutive decision cycles.

9. An SVG follow-network adaptive switching control system based on the operating short-circuit ratio, used to execute the SVG follow-network adaptive switching control method based on the operating short-circuit ratio as described in any one of claims 1-8, characterized in that, include: The measurement layer includes phasor measurement units deployed at grid connection points for real-time acquisition of voltage and current phasor data, and a new energy power station monitoring system for acquiring real-time power output data of each unit and predicted power output data of new energy sources. The computing layer includes a short-circuit ratio calculation unit, a network quantity decision unit, and a coordination control unit. The short-circuit ratio calculation unit is used for multi-timescale short-circuit ratio calculation, the network quantity decision unit is used for performing SVG target quantity calculation and switching trigger decision, and the coordination control unit is used for performing target SVG selection and remaining SVG coordination control. The execution layer consists of a cluster of multiple SVGs capable of operating in both network-building and network-following modes, with each SVG supporting smooth switching between the two control modes. The communication layer is a high-speed communication network connecting the measurement layer, the computing layer, and the execution layer, supporting millisecond-level data transmission.

Images