Voltage regulator and SVG coordinated voltage stabilizing system for photovoltaic power station

Abstract

The application discloses a voltage regulator and SVG cooperative voltage stabilizing system for a photovoltaic power station, which comprises a power grid intensity detection module for sensing the state of a power grid in real time, a network configuration type control module for dynamically adjusting virtual inertia and damping parameters of the voltage regulator according to a short-circuit ratio, so that the voltage regulator is operated in a network configuration type mode. A cross-device power distribution module integrates the power grid intensity and the device load rate, calculates active and reactive power distribution ratios, and commands the static var generator to flexibly switch between the reactive power compensation and active power absorption / release modes, so as to realize power cross-device self-balancing. A green energy-saving optimization module performs multi-objective dynamic optimization on control parameters, and a fault cooperative processing module coordinates the timing actions of the two devices in the voltage drop and recovery process. The application forms a closed-loop cooperative control of "sensing-network configuration-self-balancing-optimization", and significantly improves the voltage support capability, operation economy and fault ride-through reliability of the photovoltaic power station under a weak power grid.

Description

Technical Field

[0001] This invention relates to the field of power system control technology, and in particular to a voltage regulator and SVG coordinated voltage stabilization system for photovoltaic power plants. Background Technology

[0002] With the continuous growth of installed capacity of new energy power generation such as photovoltaics, the penetration rate of photovoltaic power plants in the power system is constantly increasing, and their grid-connected operation places higher demands on the voltage stability of the power grid. Especially in application scenarios with relatively weak grid structure and low equivalent short-circuit capacity, insufficient grid inertia and damping level limit the system's ability to withstand disturbances. Fluctuations in photovoltaic output, load changes, or grid operating condition switching may cause problems such as voltage deviation, slow recovery, or even oscillations, which has become one of the important factors restricting the high proportion of new energy integration.

[0003] In photovoltaic power plant engineering practice, grid-connected voltage regulation is typically achieved through the cooperation of voltage regulating devices and reactive power compensation devices. However, in complex power grid environments, especially when grid strength dynamically changes with operating conditions, existing voltage stabilization methods still have shortcomings in terms of overall coordination and adaptability. On the one hand, the system's ability to sense the grid's operating status is limited, making it difficult to reflect the impact of changes in grid strength on voltage regulation strategies in a timely manner, resulting in a deviation between the regulation response and the actual needs of the grid. On the other hand, during the coordinated operation of multiple types of voltage regulation equipment, there is a lack of a unified dynamic control mechanism. The power sharing relationship between the equipment is relatively fixed, making it difficult to flexibly adjust according to real-time operating conditions, which can easily lead to problems such as excessive local equipment load and decreased regulation efficiency.

[0004] Furthermore, the dynamic response characteristics of the voltage stabilization system have a significant impact on system safety when the grid voltage experiences disturbances or short-term anomalies. If the regulation process lacks sufficient equivalent inertia and damping support, shocks or secondary fluctuations are likely to occur during voltage recovery, which is detrimental to the continuous and stable operation of the photovoltaic power station. Meanwhile, with the large-scale development of new energy power stations, control methods that solely aim for voltage compliance are no longer sufficient to meet actual engineering needs. Factors such as equipment operating losses, energy efficiency levels, and long-term reliability are gradually becoming constraints that cannot be ignored in system design.

[0005] Therefore, how to effectively perceive the characteristics of the power grid without relying on additional complex configurations, and how to establish a more flexible, coordinated and adaptive voltage regulation mechanism among various voltage regulation devices so that the system can take into account voltage stability, operational stability and energy efficiency under different power grid strength conditions, has become a technical problem that urgently needs to be solved in the field of grid-connected operation of photovoltaic power plants. Summary of the Invention

[0006] To address the shortcomings of existing technologies, the present invention aims to provide a voltage regulator and SVG coordinated voltage stabilization system for photovoltaic power plants. This system enables real-time sensing of grid strength, adaptive support for grid configuration, flexible power allocation across devices, and multi-objective optimization of operational energy efficiency through closed-loop coordinated control, thereby comprehensively improving the system's voltage stability, dynamic response quality, and economic operation level.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a voltage regulator and SVG coordinated voltage stabilization system for photovoltaic power plants, comprising: The power grid strength detection module is used to collect power grid parameters in real time and calculate the real-time short-circuit ratio to determine the power grid strength level. A grid-type control module, connected to the power grid strength detection module, is used to generate a virtual inertia coefficient and a damping coefficient based on the real-time short-circuit ratio using a virtual inertia adaptive formula. A voltage regulator, connected to the grid-type control module, is used to perform dynamic tap adjustment in a grid-type mode according to the virtual inertia coefficient and the damping coefficient. The cross-device power distribution module is connected to the power grid strength detection module and the grid-type control module. It is used to obtain the real-time load rate of the voltage regulator and the real-time load rate of the static var generator. Based on the power grid strength level, the real-time short-circuit ratio, the real-time load rate of the voltage regulator and the real-time load rate of the static var generator, the active power distribution ratio and the reactive power distribution ratio are calculated. A static var generator is connected to the cross-device power distribution module and is used to switch between reactive power compensation and active power absorption / release according to the active power distribution ratio and the reactive power distribution ratio, and to undertake the corresponding reactive power regulation task. The green energy-saving optimization module connects the network control module and the cross-device power distribution module. It is used to dynamically optimize and correct the virtual inertia coefficient, damping coefficient, active power distribution ratio and reactive power distribution ratio with system voltage deviation, total loss and equipment life loss as optimization targets, and then feeds the results back to the corresponding modules. The fault coordination processing module is connected to the power grid strength detection module, the voltage regulator and the static var generator. When the power grid voltage in the power grid parameters is detected to drop above a preset threshold, the module coordinates and controls the static var generator and the voltage regulator to perform a fault ride-through action. When the voltage recovers, the module controls the two to coordinate their actions to switch to steady-state operation without impact.

[0008] Furthermore, the power grid strength detection module includes: The first acquisition unit is used to synchronously acquire the three-phase voltage, three-phase current and frequency signals at the grid connection point at a frequency not lower than the first preset frequency. An impedance calculation unit, connected to the first acquisition unit, is used to estimate the equivalent impedance of the power grid online based on the recursive least squares method according to the three-phase voltage and three-phase current. The short-circuit ratio calculation unit, connected to the impedance calculation unit, is used to calculate the real-time short-circuit ratio based on the equivalent impedance and the rated capacity of the photovoltaic power station. The strength determination unit, connected to the short-circuit ratio calculation unit, is used to compare the real-time short-circuit ratio with a first strength threshold to determine the power grid strength level; wherein, when the real-time short-circuit ratio is greater than the first strength threshold, it is determined to be a strong power grid, otherwise it is determined to be a weak power grid. The data verification and buffering unit is connected to the first acquisition unit and the impedance calculation unit respectively. It is used to verify the validity of the acquired three-phase voltage and three-phase current, and to buffer the verified data using a sliding time window to form a time-series data sequence for calculation. The impedance calculation unit is further used to perform online estimation of the equivalent impedance based on the time-series data sequence by using a recursive least squares method with a forgetting factor.

[0009] Furthermore, the network-type control module includes: A parameter receiving unit is used to receive the real-time short-circuit ratio from the power grid strength detection module; A virtual inertia calculation unit, connected to the parameter receiving unit, is used to calculate the virtual inertia coefficient based on the real-time short-circuit ratio and a preset adaptive formula, and to perform boundary constraint processing on the virtual inertia coefficient to output the final virtual inertia coefficient; wherein, the adaptive formula is configured such that when the real-time short-circuit ratio is less than or equal to a first short-circuit threshold, the virtual inertia coefficient increases as the real-time short-circuit ratio decreases; The damping coefficient calculation unit is connected to the parameter receiving unit and the virtual inertia calculation unit, respectively, and is used to calculate the damping coefficient that matches the current power grid strength level and virtual inertia based on the virtual inertia coefficient and the real-time short-circuit ratio through a dynamic matching algorithm. The output and update unit is connected to the virtual inertia calculation unit and the damping coefficient calculation unit, respectively. It is used to send the virtual inertia coefficient and the damping coefficient to the voltage regulator and receive correction instructions from the green energy-saving optimization module. It updates the virtual inertia coefficient and the damping coefficient online according to the correction instructions.

[0010] Furthermore, the voltage regulator includes: A control interface unit is used to receive the virtual inertia coefficient and the damping coefficient from the network-type control module; The VSG algorithm execution unit is connected to the control interface unit. It stores a virtual synchronous generator control algorithm, which is used to simulate the rotor motion equation of the synchronous generator according to the virtual inertia coefficient and the damping coefficient, and calculate and output the virtual power angle and virtual frequency to characterize the grid synchronization state in real time. The tap control unit, connected to the VSG algorithm execution unit, is used to generate tap adjustment commands based on the virtual power angle and the virtual frequency, combined with the grid connection point voltage feedback signal, and dynamically adjust the action rate of executing the tap adjustment commands according to the virtual inertia coefficient. The collaborative interface unit, connected to the VSG algorithm execution unit, is used to receive fault crossing instructions under the coordination of the fault collaborative processing module, and adjust the control target of the VSG algorithm execution unit according to the fault crossing instructions, so as to prioritize maintaining the stability of the virtual power angle during the fault.

[0011] Furthermore, the cross-device power allocation module includes: The data fusion unit is used to receive the grid strength level and the real-time short-circuit ratio output by the grid strength detection module, and to obtain the real-time load rate of the voltage regulator and the real-time load rate of the static var generator. The allocation ratio calculation unit, connected to the data fusion unit, is used to take the real-time short-circuit ratio, the real-time load rate of the voltage regulator, and the real-time load rate of the static var generator as input variables, calculate them through a preset collaborative allocation algorithm, and output the active power allocation ratio and the reactive power allocation ratio; wherein, the collaborative allocation algorithm is configured to dynamically determine the active power allocation ratio based on the degree to which the real-time short-circuit ratio deviates from the reference value and the difference between the two load rates. The collaborative allocation algorithm includes an active power allocation strategy, which is configured as follows: When the calculated active power allocation ratio is greater than zero, the active power allocation ratio is used to indicate that the static var generator absorbs active power, and the proportion of absorbed power does not exceed a first proportion threshold; when the active power allocation ratio is less than or equal to zero, the active power allocation ratio is used to indicate that the static var generator does not undertake the task of absorbing active power.

[0012] Furthermore, the process by which the allocation ratio calculation unit executes the collaborative allocation algorithm specifically includes: The real-time short-circuit ratio is processed based on the first weighting coefficient to obtain the first component characterizing the influence of power grid strength; The difference between the real-time load rate of the voltage regulator and the real-time load rate of the static var generator is processed based on the second weighting coefficient to obtain a second component characterizing the impact of equipment load balance. The first component and the second component are combined to obtain the active power allocation ratio; and the reactive power allocation ratio is determined based on the active power allocation ratio.

[0013] Furthermore, the static var generator includes: A control command receiving unit is used to receive the active power allocation ratio and the reactive power allocation ratio from the cross-device power allocation module; The four-quadrant mode control unit is connected to the control command receiving unit and is used to control the static var generator to switch between reactive power compensation mode, active power absorption mode and active power release mode according to the active power distribution ratio, and to determine the corresponding power command. The power execution unit is connected to the four-quadrant mode control unit and is used to execute the power command and output the corresponding reactive power and active power to undertake the reactive power adjustment task determined by the reactive power allocation ratio. The fault coordination unit, connected to the power execution unit, is used to, upon receiving a priority support instruction from the fault coordination processing module, exceed the conventional constraints of the active power allocation ratio and the reactive power allocation ratio, and prioritize controlling the power execution unit to output a rated reactive power capacity not lower than a second ratio threshold, so as to quickly support the grid voltage. The four-quadrant mode control unit is specifically used for: When the active power allocation ratio is greater than zero, the static var generator is controlled to enter a mixed operation mode of active power absorption and reactive power compensation, wherein the upper limit of the absorbed active power does not exceed the product of the active power allocation ratio and a preset capacity reference value. When the active power allocation ratio is less than or equal to zero, the static var generator is controlled to enter the pure reactive power compensation mode.

[0014] Furthermore, the green energy-saving optimization module includes: The second acquisition unit is used to acquire real-time operating parameters of the system. The real-time operating parameters of the system include at least: deviation indicators characterizing voltage stability, loss indicators characterizing system efficiency, and life indicators characterizing equipment operating status. A multi-objective optimization unit, connected to the second acquisition unit, is used to perform online iterative optimization calculations on the virtual inertia coefficient and damping coefficient from the network control module, and the active power allocation ratio and reactive power allocation ratio from the cross-device power allocation module, based on a preset heuristic optimization algorithm, using the deviation index, the loss index and the lifetime index as optimization objectives, and generating a set of optimized parameter correction values. The output and feedback unit, connected to the multi-objective optimization unit, is used to send the optimized parameter correction values ​​to the network control module and the cross-device power allocation module respectively, so as to update their output parameters.

[0015] Furthermore, the optimization objective configured in the multi-objective optimization unit is to simultaneously minimize the weighted sum of the deviation index, the loss index, and the lifetime index; The heuristic optimization algorithm is configured to perform parallel search of the solution space of the virtual inertia coefficient, the damping coefficient, the active power allocation ratio, and the reactive power allocation ratio within a preset iteration period, in order to find the parameter correction value that satisfies the optimization objective. The deviation index is the deviation between the grid connection point voltage and the rated voltage; the loss index is the total active power loss of the system, including the voltage regulator loss and the static var generator loss; and the lifespan index is a comprehensive loss factor calculated based on the number of times the voltage regulator operates and the thermal load state of the static var generator.

[0016] Furthermore, the fault collaborative processing module includes: The fault detection and identification unit is connected to the power grid strength detection module and is used to determine in real time whether a voltage drop event has occurred based on the power grid parameters, and to generate a fault trigger signal when the voltage drop exceeds a first amplitude threshold. A multi-stage instruction generation unit, connected to the fault detection and identification unit, is used to generate and output multi-stage collaborative control instructions in response to the fault trigger signal according to a preset timing strategy. The collaborative control instructions include a first control instruction, a second control instruction, a third control instruction, a fourth control instruction, and a fifth control instruction. The timing strategy includes: In the first stage, a first control command is sent to the static var generator to make it prioritize outputting rated reactive power not lower than a first proportional threshold; at the same time, a second control command is sent to the voltage regulator to make it maintain virtual power angle stability. In the second stage, a third control command is sent to the static var generator to enable it to absorb active power at its rated capacity, not exceeding the second proportional threshold, based on the redundant active power calculated by the cross-device power allocation module. During the fault recovery phase, when the voltage is detected to have recovered to above the second amplitude threshold, a fourth control command is sent to the static var generator to control it to release the absorbed active power at a rate not greater than a preset rate, and a fifth control command is simultaneously sent to the voltage regulator to control it to adjust the tap position.

[0017] The beneficial effects of this invention are: Compared with the prior art, the present invention has at least the following beneficial effects: 1. Significantly enhanced grid adaptability: This invention achieves real-time sensing of grid operating parameters and dynamically adjusts the system's equivalent inertia and damping characteristics by combining grid-based control methods. This enables the voltage regulation system to adaptively adjust its control strategy according to changes in grid strength. Even in situations where the grid structure is relatively weak or operating conditions change frequently, the voltage regulation process can still maintain stability and speed, thereby significantly improving the grid-connection adaptability of photovoltaic power plants in weak grid and complex grid environments.

[0018] 2. Reduce dependence on external energy support units: By establishing a cross-equipment power coordination mechanism between voltage regulating equipment and reactive power regulating equipment, the system can flexibly switch between different power regulation modes during voltage regulation, effectively undertaking the power regulation tasks required for voltage support. This allows for effective suppression of grid voltage disturbances without the need for additional independent energy storage devices, which helps reduce the overall system configuration complexity and construction costs.

[0019] 3. Achieving a synergistic balance between voltage stability and energy efficiency optimization: This invention introduces an optimization mechanism that comprehensively considers voltage deviation, system losses, and equipment operating status. It dynamically corrects key control parameters and power distribution relationships, effectively reducing unnecessary adjustment actions and energy losses while ensuring voltage stability. This lowers the energy consumption level of equipment during long-term operation, improves the overall operating efficiency of the system, and meets the green and low-carbon development needs of new energy systems.

[0020] 4. Improved system robustness and applicability: By comprehensively judging the grid strength level and equipment operating load status, this invention can maintain stable operation under a wide range of grid conditions and photovoltaic power output fluctuation scenarios, and has good adaptability to different grid connection environments. This is conducive to improving the versatility and reliability of photovoltaic power plant voltage stabilization systems in multiple scenarios.

[0021] 5. It helps extend the service life of key equipment: This invention dynamically balances the power load relationship between voltage regulating equipment and reactive power regulating equipment, avoiding a single equipment from being in a high-load or frequent operation state for a long time, thereby slowing down the performance degradation of the equipment, which helps extend the service life of the equipment, reduce the frequency of maintenance and long-term operating costs. Attached Figure Description

[0022] Figure 1 This is a schematic diagram of the structure of the voltage regulator and SVG coordinated voltage stabilization system used in photovoltaic power plants in this invention.

[0023] Figure reference numerals: 1. Power grid strength detection module; 11. First acquisition unit; 12. Impedance calculation unit; 13. Short-circuit ratio calculation unit; 14. Strength judgment unit; 15. Data verification and buffering unit; 2. Network-type control module; 21. Parameter receiving unit; 22. Virtual inertia calculation unit; 23. Damping coefficient calculation unit; 24. Output and update unit; 3. Voltage regulator; 31. Control interface unit; 32. VSG algorithm execution unit; 33. Tap control unit; 34. Coordination interface unit; 4. Cross-device power allocation module; 41. Data fusion unit; 42. Allocation ratio calculation unit; 5. Static var generator; 51. Control command receiving unit; 52. Four-quadrant mode control unit; 53. Power execution unit; 54. Fault coordination unit; 6. Green energy-saving optimization module; 61. Second acquisition unit; 62. Multi-objective optimization unit; 63. Output and feedback unit; 7. Fault coordination processing module; 71. Fault detection and identification unit; 72. Multi-stage command generation unit. Detailed Implementation

[0024] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Identical components are denoted by the same reference numerals. It should be noted that the terms "front," "rear," "left," "right," "upper," and "lower" used in the following description refer to directions in the accompanying drawings, and the terms "bottom surface," "top surface," "inner," and "outer" refer to directions toward or away from the geometric center of a specific component, respectively.

[0025] Example 1, refer to Figure 1 This is the first embodiment of the present invention. This embodiment provides a voltage regulator 3 and SVG coordinated voltage stabilization system for photovoltaic power plants. It can realize real-time sensing of grid strength, adaptive support of grid structure, flexible power allocation across devices and multi-objective optimization of operating energy efficiency through closed-loop coordinated control, thereby comprehensively improving the voltage stability, dynamic response quality and economic operation level of the system.

[0026] I. Implementation Environment and Overall Structure Description; This embodiment is applicable to grid-connected operation scenarios of centralized or distributed photovoltaic power plants, especially suitable for weak grids or grids with fluctuating short-circuit ratios (SCRs) ranging from 0.8 to 3.0. The system is deployed at the grid connection point (PCC) of the photovoltaic power plant and is directly electrically connected to the upstream grid for dynamic adjustment and stabilization control of the grid-connected voltage.

[0027] The voltage regulator 3 and SVG coordinated voltage stabilization system in this embodiment includes: The system comprises a power grid strength detection module 1, a grid-type control module 2, a voltage regulator 3, a cross-equipment power distribution module 4, a static var generator 5, a green energy-saving optimization module 6, and a fault collaborative processing module 7. Each module interacts with the other via industrial Ethernet and Profibus-DP dual redundant communication bus, and is coordinated and controlled by a unified controller.

[0028] II. Hardware Components and Actual Selection; 1. Voltage regulator 3; In this embodiment, the voltage regulator 3 is an on-load tap-changing transformer with a rated capacity of 500kVA, a tap adjustment step size of 0.5%, and a single tap adjustment response time of no more than 50ms.

[0029] A grid-type control unit based on the DSP chip TMS320F28335 is integrated inside the voltage regulator 3 to implement the virtual synchronous generator (VSG) control algorithm, enabling the voltage regulator 3 to have grid-type operation capability.

[0030] The voltage regulator 3 not only performs the traditional voltage level regulation function in the system, but also participates in the dynamic stability process of the system through equivalent inertia and damping regulation in the grid-type control mode.

[0031] 2. Static Var Generator 5 (SVG); The static var generator 5 adopts a four-quadrant operation type structure, with a rated capacity of 300kVA, and has the following capabilities: Reactive power adjustment range: -100% to +100% of rated capacity; Active power absorption / release range: 0–30% of rated capacity; The dynamic response time is no more than 20ms.

[0032] The static var generator 5 can dynamically switch between reactive power compensation mode and "reactive power compensation + active power absorption / release" mode through the control of the cross-device power distribution module 4.

[0033] 3. Power grid strength detection module 1; This module consists of the following sensors: voltage sensor: accuracy class 0.2; current sensor: accuracy class 0.2; frequency sensor: measurement range 45~55Hz, accuracy class 0.01Hz.

[0034] The sampling frequency is set to 1kHz to collect voltage, current and frequency data at the grid connection point in real time, providing basic data for subsequent grid strength assessment.

[0035] 4. Controller and communication environment; The system uses an industrial-grade PLC (Siemens S7-1500) as the main controller, responsible for: determining the power grid strength level; calculating cross-device power allocation; running optimization algorithms; and scheduling fault coordination logic.

[0036] The PLC's operation cycle is no more than 10ms. Communication adopts a dual-bus redundancy design of Ethernet and Profibus-DP, with a communication delay of no more than 10ms, to ensure the real-time performance and reliability of control commands.

[0037] III. Software Functional Modules and Model Implementation; 1. Working principle of power grid strength detection module 1; Based on real-time acquired voltage and current signals, this module extracts the fundamental component through Fourier transform and calculates the real-time short-circuit ratio (SCR) of the power grid. The real-time SCR is updated every 10ms and used as the basis for determining the power grid strength level: when SCR > 1.5, it is determined to be a relatively strong power grid; when SCR ≤ 1.5, it is determined to be a weak power grid.

[0038] Technical benefits: By quantifying the strength of the power grid, the system control strategy can be adaptively adjusted based on the real-time characteristics of the power grid, avoiding the problem of voltage oscillation caused by traditional fixed parameter control under weak power grid conditions.

[0039] 2. Working principle of the network-type control module 2; The network control module 2 operates based on the VSG model and adjusts the dynamic behavior of the voltage regulator 3 through the virtual inertia coefficient J and damping coefficient D.

[0040] In this embodiment, the values ​​of the virtual inertia coefficient J and the damping coefficient D are adjusted according to the SCR through the following adaptive relationship: the adjustment range of the virtual inertia coefficient J is 0.5s to 1.2s; the adjustment range of D is 0.8 to 2.4.

[0041] When the real-time short-circuit ratio (SCR) of the power grid decreases, the grid-type control module 2 automatically increases the values ​​of the virtual inertia coefficient and damping coefficient, so that the voltage regulator 3 exhibits a larger equivalent inertia and damping in the system.

[0042] Technical effect: Through grid-type control, the voltage regulator 3 is transformed from a traditional passive voltage regulator into a "voltage source" unit that actively participates in system stability, which significantly suppresses voltage fluctuations and improves voltage recovery speed under weak grid conditions.

[0043] 3. Working principle of cross-device power distribution module 4; This module is deployed in the PLC and is used to obtain: the real-time load rate of voltage regulator 3; the real-time load rate of SVG; and the current power grid strength level and the real-time short-circuit ratio of the power grid.

[0044] Based on this, the cross-device power allocation module 4 calculates the active power allocation ratio and reactive power allocation ratio according to the preset model and updates it every 10ms.

[0045] Technical effect: It breaks the fixed division of labor between voltage regulator 3 (which only regulates voltage) and static var generator 5 (which only supplements reactive power), enabling the two types of equipment to dynamically share the adjustment task according to the grid status and their own load, thereby achieving equipment load balance and reducing system losses.

[0046] 4. Working principle of green energy-saving optimization module 6; The Green Energy Saving Optimization Module 6 uses an embedded improved particle swarm optimization algorithm as its core, and its optimization objectives include: system voltage deviation; total system loss and equipment lifespan-related losses.

[0047] The algorithm parameters are set as follows: population size 30; number of iterations 50; optimization calculation is performed every 500ms.

[0048] The optimization results are used to dynamically correct the virtual inertia coefficient, active power allocation ratio, and reactive power allocation ratio, and are fed back to the network control module 2 and the cross-device power allocation module 4.

[0049] Technical benefits: While ensuring voltage stability, unnecessary tapping operations and high-frequency switching of the static var generator 5 are reduced, achieving a synergistic balance between voltage control objectives and energy saving and life extension objectives.

[0050] 5. Working principle of fault collaborative processing module 7; The fault collaborative processing module 7 monitors the grid voltage in real time. When it detects a voltage drop exceeding 10%, it immediately triggers the fault handling process.

[0051] During a fault: Static var generator 5 rapidly injects reactive power and simultaneously absorbs some active power; voltage regulator 3 suppresses voltage surges under grid-type control; after the fault is cleared, both switch to steady-state operation without impact according to a preset slope. This module ensures consistent action timing through a dual mechanism of hard-wired triggering and software commands.

[0052] Technical benefits: Avoids voltage surges or secondary fluctuations during voltage recovery, improving system stability and safety in the event of power grid failures.

[0053] IV. Overall technical effects of Example 1; During system operation, the system goes through the following stages in sequence: initialization, power grid strength detection, normal operation, weak grid operation, fault handling, and dynamic adjustment.

[0054] Through the synergistic effect of the above modules, this embodiment can achieve: The grid-connected voltage fluctuation range is controlled within ±1%; the voltage recovery time is no more than 100ms; the overall energy consumption of the system is reduced by no less than 15%; the number of taps of the voltage regulator is reduced by no less than 30%; it is suitable for grid environments with a real-time short-circuit ratio of 0.8 to 3.0; and it is adaptable to dynamic fluctuations of ±20% in photovoltaic output.

[0055] Example 2 is the second embodiment of the present invention. This embodiment is a preferred implementation based on Example 1 above, which further refines and expands the accuracy of power grid strength detection, the logic for generating grid-type control parameters, and the internal control structure of voltage regulator 3.

[0056] I. Overall description and applicable scenarios of Example 2; This embodiment is particularly suitable for grid-connected photovoltaic power plants with frequent changes in grid impedance at the grid connection point and significant dynamic fluctuations in the short-circuit ratio (SCR). It can effectively improve the voltage stability and dynamic response capability of the system under weak grid conditions.

[0057] The voltage regulator 3 and SVG coordinated voltage stabilization system in this embodiment still includes voltage regulator 3, static var generator 5, grid strength detection module 1, grid-type control module 2, cross-device power distribution module 4, green energy-saving optimization module 6, and fault coordination processing module 7. The modules interact with each other through a high-speed communication bus, and the communication delay is no more than 10ms.

[0058] II. Structure and working principle of power grid strength detection module 1; 1. Module structure composition; In this embodiment, the power grid strength detection module 1 includes the following functional units: 1) The first acquisition unit 11 is used to synchronously acquire the three-phase voltage, three-phase current and frequency signals at the grid connection point at a frequency not lower than the first preset frequency.

[0059] In this embodiment, the first preset frequency is explicitly set to 1kHz to ensure the timeliness and accuracy of subsequent online impedance estimation and short-circuit ratio calculation.

[0060] 2) The data verification and buffering unit 15 is connected to the first acquisition unit 11 and the impedance calculation unit 12 respectively. It is used to verify the validity of the acquired three-phase voltage and three-phase current data, remove abnormal sampling points, and use a sliding time window method to buffer the verified data to form a continuous and stable time-series data sequence.

[0061] 3) Impedance calculation unit 12, connected to the first acquisition unit 11 and the data verification and buffer unit 15, is used to estimate the equivalent impedance of the power grid at the grid connection point online based on the time-series data sequence and using a recursive least squares method incorporating a forgetting factor. The online estimation process specifically includes: At the grid connection point, the power grid is equivalent to a linear impedance model, and its instantaneous voltage and current relationship is expressed as: ; in, Let be the fundamental component of the grid-connected voltage at the k-th sampling time. Let K be the fundamental component of the grid-connected current at the k-th sampling time. This is the Hilbert transform operator, used to generate virtual components orthogonal to the current. The equivalent resistance parameters to be estimated are: The equivalent reactance parameters to be estimated are: This is the modeling error term; The above formula, by introducing the Hilbert transform, allows the resistance and reactance parameters to be identified simultaneously within the same linear regression framework, thereby improving the observability of the model.

[0062] Then construct the parameter vector With regression vector : ; The recursive update process is as follows: Gain vector calculation formula: ; in, For the parameter gain vector, Estimate the covariance matrix of the parameters at the k-th sampling time. The forgetting factor has a value range of 0.95 ≤ <1 is used to reduce the weight of historical data and enhance the ability to track new operating conditions.

[0063] Parameter update formula: ; This formula introduces the hyperbolic tangent function tanh(⋅): used to nonlinearly suppress abnormal measurement errors; To prevent sudden noise from causing drastic parameter jumps; to improve the stability and robustness of impedance estimation.

[0064] Covariance matrix update formula: ; In the above formula, The smaller the value, the more stable and reliable the parameter estimation. A larger value indicates that the power grid is in a period of rapid change or that the measurement noise is high. This can be addressed by using a forgetting factor. ,make It can quickly reflect actual operating conditions such as changes in power grid topology and the switching on and off of grid-connected equipment.

[0065] Technical effects: Through the above online estimation mechanism, the equivalent impedance of the power grid can be updated in real time at the millisecond level; the accuracy of short-circuit ratio calculation is significantly improved; and a real and dynamic power grid strength input is provided for the grid-type control module 2, solving the parameter mismatch problem in weak power grid scenarios.

[0066] 4) The short-circuit ratio calculation unit 13 is connected to the impedance calculation unit 12. It is used to calculate the real-time short-circuit ratio (SCR) based on the equivalent impedance obtained by online estimation and the rated capacity of the photovoltaic power station.

[0067] The formula for calculating the real-time short-circuit ratio (SCR) is: ; in, Rated voltage at the grid connection point; SCR is the real-time short-circuit ratio of the power grid; The rated capacity of the photovoltaic power station. for Given the equivalent resistance and reactance, The equivalent impedance magnitude of the power grid. Let be the matrix trace function, representing the overall estimation uncertainty; It is a logarithmic function used to suppress the amplification effect of covariance anomalous; This is the uncertainty weighting coefficient, used to adjust the degree of influence of covariance on SCR.

[0068] Using the SCR update formula above: When the impedance estimation has not yet converged ( (Larger): SCR is automatically "reduced", the system is more inclined to judge it as a weak grid, and enters a conservative grid control strategy in advance to avoid control mismatch.

[0069] When the impedance estimation stabilizes ( (Smaller): SCR is closer to the actual power grid strength, improving the accuracy of control parameter matching.

[0070] To avoid frequent fluctuations in the determination of power grid strength: the combined effect of logarithmic function and trace operation makes the SCR change continuous and smooth, significantly improving the system robustness.

[0071] 5) Strength determination unit 14: The short-circuit ratio calculation unit 13 is connected to compare the real-time short-circuit ratio with a first intensity threshold to determine the power grid intensity level. In this embodiment, the first intensity threshold is explicitly set to 1.5. When the real-time short-circuit ratio is greater than 1.5, it is determined to be a strong power grid; otherwise, it is determined to be a weak power grid.

[0072] 2. Working principle and technical effects; Through the above structure, the power grid strength detection module 1 can realize continuous online estimation of power grid impedance and short-circuit ratio, avoiding the error accumulation problem of traditional static parameter-based or offline calculation, enabling the system to perceive the change process of the power grid from strong to weak or from weak to strong in real time and providing a reliable input basis for grid-type control.

[0073] III. Structure and parameter generation mechanism of network-type control module 2; 1. Module structure composition; Network-type control module 2 includes: Parameter receiving unit 21 is used to receive the real-time short-circuit ratio (SCR) from the power grid strength detection module 1; Virtual inertia calculation unit 22 is used to calculate virtual inertia coefficient based on real-time short-circuit ratio (SCR). Damping coefficient calculation unit 23 is used to generate a damping coefficient that matches the virtual inertia and power grid strength level; The output and update unit 24 is used to send the generated virtual inertia coefficient and damping coefficient to the voltage regulator 3, and to update them online according to the correction instructions of the green energy-saving optimization module 6.

[0074] 2. The adaptive generation principle of virtual inertia coefficient; The virtual inertia calculation unit 22 generates virtual inertia coefficients based on a preset adaptive formula and performs boundary constraint processing on them to obtain the final virtual inertia coefficients.

[0075] The virtual inertia coefficient is calculated using a composite model constructed with the Sigmoid function and the exponential enhancement function, as shown in the formula: ; Where J is the virtual inertia coefficient. The baseline virtual inertia coefficient is set to 0.5s in this embodiment; The maximum virtual inertia limit is set to 1.2s in this embodiment; SCR is the real-time short-circuit ratio. The sigmoid steepness coefficient is used to control the smoothness of the transition between strong and weak power grids; The weighting coefficients are used to enhance inertia support near critical weak power grids.

[0076] Technical effects: The Sigmoid function ensures the continuity and differentiability of inertia as it changes in real time, avoiding abrupt changes in control; the exponential enhancement term provides additional inertia support around the real-time short-circuit ratio SCR≈1.5, significantly improving the stability of weak power grids.

[0077] The adaptive logic is configured such that when the real-time short-circuit ratio (SCR) is less than or equal to the first short-circuit threshold, the virtual inertia coefficient increases as the real-time short-circuit ratio (SCR) decreases.

[0078] In this embodiment, the first short-circuit threshold and the first intensity threshold are the same, both set to 1.5.

[0079] Meanwhile, the virtual inertia coefficient is limited to a preset safe range through a boundary restriction mechanism to avoid excessive inertia causing slow system response.

[0080] 3. The principle of dynamic matching of damping coefficient; The damping coefficient calculation unit 23 generates the damping coefficient through a dynamic matching algorithm based on the current virtual inertia coefficient and the real-time short-circuit ratio SCR, so that the system can maintain a stable operating state under different power grid strength and inertia levels.

[0081] The damping coefficient D is dynamically matched using a hyperbolic tangent function plus a logarithmic function, as shown in the formula: ; Where D is the damping coefficient. Here, c is the damping reference coefficient; c and d are both nonlinear adjustment factors. tanh(⋅) is the hyperbolic tangent function, used to suppress excessive damping amplification under high inertia.

[0082] Technical effect: Achieve dynamic balance between "high inertia without high damping and low inertia without low damping", ensuring that the grid-type control maintains good damping characteristics within the real-time short-circuit ratio (SCR) range.

[0083] 4. Technical effects; Through the structure and logic of the grid-type control module 2, the voltage regulator 3 has higher equivalent inertia and reasonable damping under weak grid conditions, thereby improving the system's ability to suppress voltage disturbances; under strong grid conditions, unnecessary inertia amplification is avoided, ensuring the system's response speed.

[0084] IV. Internal structure and network operation mechanism of voltage regulator 3; 1. Internal structure of voltage regulator 3; The voltage regulator 3 in this embodiment includes: The control interface unit 31 is used to receive the virtual inertia coefficient and damping coefficient from the network control module 2; VSG algorithm execution unit 32 internally stores the virtual synchronous generator control algorithm; The tap control unit 33 is used to perform the tap adjustment action of the voltage regulator 3; The collaborative interface unit 34 is used to exchange information with the fault collaborative processing module 7.

[0085] 2. The principle of VSG algorithm execution unit 32; The VSG algorithm execution unit 32 simulates the rotor motion equation of the synchronous generator based on the received virtual inertia coefficient and damping coefficient, and calculates the virtual power angle and virtual frequency in real time to characterize the synchronization state between the voltage regulator 3 and the power grid.

[0086] The VSG algorithm execution unit 32 adopts an improved nonlinear rotor model, the formula of which is: ; ; ; in, For virtual angular velocity, For virtual power angle, This is the equivalent mechanical input power command. The equivalent electromagnetic power is calculated based on the grid connection point voltage and current. This is the power angle coupling coefficient, used to suppress large power angle oscillations. For a nonlinear sinusoidal function, a synchronization stability constraint is introduced. The angular frequency is the rated synchronization frequency, and τ is a continuous time variable. Let τ be the virtual angular velocity (i.e., the instantaneous value of the virtual frequency) of the virtual synchronous generator at the time corresponding to the integral time variable τ.

[0087] The virtual power angle reflects the phase relationship between the equivalent voltage source of the voltage regulator 3 and the power grid, and its stability directly affects the smooth transmission of active power.

[0088] Based on the above model formula, the voltage regulator 3 has the characteristics of a "virtual rotating body" at the control level; it can actively provide inertia and damping support under weak power grid conditions, and at the same time effectively suppress low-frequency oscillations of voltage and power.

[0089] 3. Tap adjustment and dynamic rate control; The tap control unit 33 generates tap adjustment commands based on the virtual power angle, virtual frequency, and grid connection point voltage feedback signals, and dynamically adjusts the tap adjustment speed according to the virtual inertia coefficient, which is achieved through the following calculation process: The target adjustment amount for tap adjustment is constructed. The tap control unit 33 first constructs the comprehensive adjustment error: ; in, To comprehensively adjust the error, The Sigmoid function is used to smooth out the effects of the angle of attack. The voltage at the grid connection point. , All of these are preset weighting coefficients.

[0090] Adaptive adjustment of tapping action rate, defined as: ; in, The current tapping rate, and These represent the minimum and maximum tap speeds, respectively. This is the sensitivity coefficient for rate adjustment. The larger the virtual inertia, the smoother the tap movement, directly reflecting the constraint of the network control on the mechanical motion layer.

[0091] Tap adjustment instructions are generated, and the final tap adjustment instructions are expressed as follows: ; in, To change the level of the tap, To control the cycle, This is a rounding function that matches the actual mechanical tap pitch.

[0092] Through the above control process: tap regulation no longer depends solely on voltage error; the operation of voltage regulator 3 is deeply coupled with the dynamic state of the power grid; and the risk of tap jitter and over-regulation is significantly reduced under weak power grid and disturbance conditions.

[0093] 4. Control mechanism during fault coordination; Under the coordination of the fault collaborative processing module 7, the collaborative interface unit 34 receives the fault ride-through command, and the voltage regulator 3 adjusts the control target of the VSG algorithm accordingly to prioritize the stability of the virtual power angle during the fault period and avoid secondary disturbances caused by frequent tapping.

[0094] 5. Technical effects; Through the above structure and operating mechanism, the voltage regulator 3 is transformed from a traditional passive voltage regulating device into an active voltage support unit with grid construction capability, which significantly improves voltage stability and system operation stability under weak grid and fault conditions.

[0095] V. Overall technical effects of Example 2; Compared to Embodiment 1, this embodiment further refines the internal structure of the power grid strength detection module 1, the grid-type control module 2, and the voltage regulator 3, enabling the system to have more accurate power grid status perception capabilities and more stable grid-type control characteristics. Under weak power grid conditions, it effectively reduces voltage fluctuation amplitude, shortens voltage recovery time, and provides a more reliable control basis for subsequent cross-device power allocation and energy-saving optimization.

[0096] Example 3 is the third embodiment of the present invention. Unlike the previous embodiment, this embodiment is a preferred implementation that further refines the power coordination and distribution mechanism between the voltage regulator 3 and the static var generator 5 and the four-quadrant operation control mode of the static var generator 5 based on Example 1 and Example 2.

[0097] I. General description of Example 3; This embodiment is particularly suitable for grid-connected operation scenarios where photovoltaic output fluctuates greatly and grid intensity changes frequently. By dynamically allocating active and reactive power regulation tasks, it achieves a comprehensive improvement in system voltage stability, equipment load balance, and operational economy.

[0098] In this embodiment, the system still includes a voltage regulator 3, a static var generator 5, a grid strength detection module 1, a grid-type control module 2, a cross-device power distribution module 4, a green energy-saving optimization module 6, and a fault collaborative processing module 7. The communication methods, operating environment, and basic parameter settings between each module are consistent with those in the aforementioned embodiment.

[0099] II. Structure and working principle of cross-device power distribution module 4; 1. Module structure composition; The cross-device power allocation module 4 in this embodiment includes: 1) Data fusion unit 41; It is used to receive the grid strength level and real-time short-circuit ratio (SCR) output by the grid strength detection module 1, and simultaneously obtain the real-time load rate of the voltage regulator 3 and the real-time load rate of the static var generator 5.

[0100] The load factor is used to characterize the proportion of power currently being borne by each device relative to its rated capacity.

[0101] 2) Allocation ratio calculation unit 42; The data fusion unit 41 is used to take the real-time short-circuit ratio (SCR), the real-time load rate of the voltage regulator 3, and the real-time load rate of the static var generator 5 as input variables, calculate them through a preset collaborative allocation algorithm, and output the active power allocation ratio and the reactive power allocation ratio.

[0102] 2. The core logic of the collaborative allocation algorithm; The collaborative allocation algorithm is configured as follows: Taking into account the degree to which the power grid intensity deviates from the benchmark value and the difference in load rate between voltage regulator 3 and static var generator 5, the active power allocation ratio is dynamically determined.

[0103] In the specific implementation process, the allocation ratio calculation unit 42 performs the following steps: The real-time short-circuit ratio (SCR) is processed based on the first weighting coefficient to obtain the first component characterizing the influence of grid strength. The difference between the real-time load rate of voltage regulator 3 and the real-time load rate of static var generator 5 is processed based on the second weighting coefficient to obtain the second component characterizing the impact of equipment load balance. The first component and the second component are combined to generate the active power allocation ratio, and the reactive power allocation ratio is determined based on the active power allocation ratio.

[0104] The active power allocation ratio is calculated using a composite expression of a normalized exponential function and a hyperbolic function, as shown in the formula: ; in, For the proportion of active power allocation, As the first weighting coefficient, This is the second weighting coefficient. For the real-time load rate of the voltage regulator, For SVG real-time load rate, As the first component, This is the second component.

[0105] when When the value is greater than 0, the SVG absorbs active power, and its upper limit of absorption is: ; in, The sigmoid function is used for soft-limit absorbing boundaries. The range of values ​​is , This represents the actual active power absorbed. This is the baseline value for the rated active capacity of the SVG.

[0106] Technical benefits: The above formula can prevent the active power distribution ratio from jumping under critical operating conditions, thereby improving the continuity of power distribution and equipment safety.

[0107] The formula for reactive power allocation ratio is: ; in, To determine the reactive power distribution ratio, a hyperbolic function is used to limit the rate of change of reactive power distribution, ensuring the smoothness of voltage regulation.

[0108] The reactive power allocation ratio and the active power allocation ratio are complementary to ensure the continuity of the overall reactive power regulation capability of the system.

[0109] 3. Active power allocation strategy and threshold limits; The collaborative allocation algorithm further includes an active power allocation strategy, specifically configured as follows: When the calculated active power allocation ratio When greater than zero, Used to indicate the active power absorbed by the static var generator; The proportion of active power absorbed does not exceed the first proportion threshold; in this embodiment, the first proportion threshold is explicitly set to 30% of the rated capacity of the static var generator. When the active power allocation ratio When the static var generator is less than or equal to zero, it does not undertake the task of active power absorption, but only participates in reactive power regulation.

[0110] The pivotal role of cross-device power distribution module 4 in the overall system: In this embodiment, the cross-device power allocation module 4 not only undertakes the traditional power allocation calculation function, but also exists as the core decision-making hub connecting the power grid sensing layer, the grid control layer and the power execution layer. Its internal data processing and decision-making mechanism constitutes one of the key innovations of the overall technical solution of this invention.

[0111] Specifically, this module introduces the following multi-source heterogeneous information as decision inputs simultaneously: The grid strength level and real-time short-circuit ratio (SCR) output by the grid strength detection module 1 are used to reflect the support capacity and stability margin of the external power grid. The real-time load rate of the voltage regulator 3 is used to reflect the regulating pressure currently borne by the grid-type voltage regulator 3; The real-time load rate of static var generator 5 is used to reflect its remaining regulation capacity and operating margin.

[0112] The above information comes from the grid side, the grid construction side, and the power execution side, and their information dimensions and physical meanings are different. The cross-device power allocation module 4 integrates the above multi-source information through a collaborative allocation algorithm to form a unified power regulation decision result.

[0113] "Active power allocation ratio" serves as a key driving signal for cross-device collaboration: In this embodiment, the active power allocation ratio output by the cross-device power allocation module 4 is not a simple power ratio parameter, but a core control command that drives a substantial change in the system's operating mode.

[0114] Specifically, when the active power distribution ratio is greater than zero, the ratio directly triggers the static var generator 5 to switch from the traditional "reactive power compensation only mode" to the "hybrid operation mode of active power absorption and reactive power compensation", enabling the static var generator 5 to break through its traditional functional boundaries and participate in the dynamic adjustment of the active power level. When the active power distribution ratio is less than or equal to zero, the static var generator 5 automatically reverts to the pure reactive power compensation mode to avoid introducing unnecessary energy exchange in operating conditions where active power coordination is not required.

[0115] Therefore, the active power allocation ratio constitutes the logical bridge between the control strategy of the grid-type voltage regulator 3 and the four-quadrant operation capability of the SVG, and is a necessary condition for realizing deep collaboration between the two types of equipment at the active power level.

[0116] Analysis of the irreplaceability of cross-device power allocation mechanisms: If the aforementioned cross-device power allocation module 4 and its collaborative allocation algorithm are missing, even if the system possesses the following individual capabilities: Power grid strength detection module 1 can accurately identify strong or weak power grids; Voltage regulator 3 has network-type control capability and can dynamically adjust virtual inertia; The static var generator 5 has four-quadrant operation capability; Each module can only operate with its own independent control objectives, and cannot form a unified and collaborative decision-making based on the global state, which will significantly limit its technical effectiveness.

[0117] In this embodiment, the cross-device power allocation module 4 incorporates both grid strength information and device load status into the decision-making process, enabling the system to reach a consistent decision across three levels: "whether active power coordination is needed," "who should undertake active power regulation," and "to what extent." This avoids: Voltage regulator 3 bears excessively high active power regulation pressure for a long time under weak power grid conditions; Static var generator 5 is unable to participate in active power regulation when it is capable of doing so. The two types of equipment may experience regulatory redundancy or mutual cancellation due to a lack of coordination.

[0118] Overall technical effects resulting from cross-device power distribution module 4: Based on the above decision-making mechanism, this embodiment, compared to a system without a cross-device power allocation mechanism, has at least the following further technical improvements: 1. Achieve dynamic coordination of active and reactive power regulation across devices: Through the linkage generation of active and reactive power allocation ratios, the voltage regulator 3 and the static var generator 5 are no longer limited to fixed functional divisions, but can dynamically cooperate according to the grid status.

[0119] 2. Significantly enhance voltage stability in weak grid scenarios: Under weak grid conditions, by guiding SVG to participate in active power absorption, the instantaneous regulation pressure of voltage regulator 3 is reduced, and the risk of voltage fluctuation and oscillation is reduced.

[0120] 3. Improve the robustness and adaptability of the overall system operation; the cross-device power distribution module 4 enables the system to automatically adjust the power regulation strategy under different grid strengths and different load conditions, thereby improving its adaptability to complex operating conditions.

[0121] 4. Provide a unified adjustment entry point for subsequent energy-saving optimization and fault coordination: The active power allocation ratio and reactive power allocation ratio serve as system-level decision results, providing a clear and adjustable control interface for the green energy-saving optimization module 6 and the fault coordination processing module 7, thereby enhancing the overall coordination of the system.

[0122] In summary, this embodiment introduces a multi-source information fusion and collaborative decision-making mechanism with the cross-device power distribution module 4 as the core, organically integrating grid strength sensing, grid-type voltage regulation control, and the four-quadrant operation capability of the static var generator 5 to form a unified, dynamic, and scalable voltage regulation control system. This mechanism constitutes one of the key innovations of this invention that distinguish it from existing technologies.

[0123] III. Structure and four-quadrant operation principle of static var generator 5; 1. The structural composition of the static var generator 5; The static var generator 5 in this embodiment includes: Control command receiving unit 51 is used to receive the active power allocation ratio from the cross-device power allocation module. With reactive power allocation ratio ; The four-quadrant mode control unit 52 is used to determine the active power distribution ratio. The size of the static var generator controls the switching between different operating modes and determines the corresponding power command; The power execution unit 53 is used to execute the corresponding power command and output active power and reactive power. The fault coordination unit 54 is used to receive priority support instructions in the event of a fault.

[0124] 2. Four-quadrant mode control logic; The four-quadrant mode control unit 52 allocates power according to the active power distribution ratio. The following control logic is executed based on the value of : When the active power allocation ratio When the value is greater than zero, the generated power command controls the static var generator to enter a mixed operation mode of active power absorption and reactive power compensation; In this mode, the upper limit of absorbed active power does not exceed the active power allocation ratio. The product of the preset capacity baseline value; At the same time, according to the reactive power allocation ratio It outputs the corresponding reactive power for voltage regulation.

[0125] When the active power allocation ratio When the value is less than or equal to zero, the generated power command controls the static var generator to enter pure reactive power compensation mode and does not undertake active power regulation tasks.

[0126] 3. Fault-priority support mechanism; When the fault coordination module 7 issues a priority support command, the fault coordination unit 54 causes the static var generator 5 to temporarily exceed the normal active power allocation ratio and reactive power allocation ratio constraints, and prioritizes the power execution unit 53 to output a rated reactive power capacity that is not lower than the second ratio threshold.

[0127] In this embodiment, the second proportional threshold is explicitly set to 60% of the rated reactive power capacity of the static var generator 5, which is used to quickly support the grid voltage in fault scenarios such as voltage drop.

[0128] 4. Technical effects; By combining four-quadrant operation with a fault-priority support mechanism, the static var generator 5 has flexible power regulation capabilities during normal operation and can quickly and centrally provide reactive power support in fault conditions, thereby significantly improving the system's voltage support capability and fault ride-through performance.

[0129] IV. Overall technical effects of Example 3; Through the cross-device power distribution module 4 and the four-quadrant control mechanism of the static var generator 5 in this embodiment, the system can achieve active and reactive power coordinated regulation between the voltage regulator 3 and the static var generator 5 under different grid strength and load conditions, effectively reducing voltage fluctuations, alleviating uneven equipment load, and providing fast and reliable voltage support under fault conditions, further improving the stability and safety of the photovoltaic power station's grid-connected operation.

[0130] Example 4 is the fourth embodiment of the present invention. Unlike the previous embodiment, based on the hardware and basic control structure of Examples 1 to 3, it further introduces the green energy-saving optimization module 6 and the fault collaborative processing module 7 to form a closed-loop collaborative control system covering all operating conditions, including "normal operation, disturbance operation, fault operation and recovery operation".

[0131] I. Functional Positioning; In this embodiment, the green energy-saving optimization module 6 serves as the long-term performance tuning layer of the system. With the goal of reducing energy consumption and extending equipment life, it performs online correction on the key control parameters of the grid-type control module 2 and the cross-device power distribution module 4. The fault collaborative processing module 7 serves as a rapid safety assurance layer for the system. When faults such as voltage drops occur, it temporarily takes over the control priority of the system to ensure voltage support capability and operational safety.

[0132] The two are independent in terms of control timing but coupled at the parameter level, together forming a "dual guarantee" of system stability and economy.

[0133] II. Structure and working principle of green energy-saving optimization module 6; (a) Module structure; Green energy-saving optimization module 6 includes: The second acquisition unit 61 is used to acquire real-time operating parameters of the system, including at least: voltage stability deviation index, system efficiency loss index and equipment life index.

[0134] The multi-objective optimization unit 62 is used to perform online iterative optimization of the following parameters based on a preset heuristic optimization algorithm, using deviation index, loss index, and lifetime index as multi-objective optimization objectives: The virtual inertia coefficient output by the network-type control module 2; the damping coefficient output by the network-type control module 2; the active power allocation ratio output by the cross-device power allocation module 4; and the reactive power allocation ratio output by the cross-device power allocation module 4.

[0135] The output and feedback unit 63 is used to feed back the optimized parameter correction values ​​to the network control module 2 and the cross-device power distribution module 4 respectively, forming a closed-loop parameter update.

[0136] (ii) Optimize the definition of objectives and indicators; In this embodiment, the multi-objective optimization unit 62 adopts a weighted multi-objective minimization strategy, and its optimization objective can be expressed as: Deviation index: The deviation between the real-time voltage at the grid connection point and the rated voltage, used to measure the voltage stabilization effect; Loss index: Total active power loss of the system, including copper loss and iron loss of voltage regulator 3, and switching and conduction losses of SVG; Life index: The comprehensive life loss factor is calculated based on the number of taps of the voltage regulator, the thermal load status of the SVG, and the cumulative stress level.

[0137] The three categories of indicators correspond to the system's stability, efficiency, and reliability, respectively. They are weighted to form a unified optimization target, thereby avoiding excessive equipment wear caused by solely pursuing voltage regulation performance.

[0138] (III) Optimization of calculation and parameter correction mechanism; The heuristic optimization algorithm is configured to perform a parallel search of the solution space for the virtual inertia coefficient, damping coefficient, active power allocation ratio, and reactive power allocation ratio within a preset iteration period (e.g., 500ms to 1s). Its specific working principle includes: The heuristic multi-objective optimization algorithm periodically performs joint optimization on the following control parameters during system operation: the virtual inertia coefficient J and damping coefficient D in the network control module 2; and the active power allocation ratio in the cross-device power allocation module 4. Reactive power allocation ratio .

[0139] The algorithm uses the real-time operating status of the system as feedback to continuously search for the optimal balance point among the three objectives of "stability, energy consumption and lifespan", thereby achieving green and energy-saving operation of the system.

[0140] Optimize variable and search space definitions: (a) Optimize the variable vector; Define the optimization variable vector as follows: ; (ii) Variable range constraints; ; By setting the above value range, we can ensure that: The grid-type voltage regulator 3 has sufficient but not excessive inertia and damping; The active power contribution of SVG does not exceed the equipment capacity; The reactive power distribution ratio is always physically feasible.

[0141] In each optimization cycle, the current system operating status is collected to generate multiple sets of candidate parameter combinations, and then the system response under each candidate parameter combination is predicted and evaluated; the parameter combination that satisfies the minimization of the weighted objective function is selected as the optimal solution; the corresponding parameter correction value is output and fed back to the relevant modules.

[0142] In this way, the system can gradually evolve towards a low-loss, low-operation-frequency, and low-thermal-stress operating state while meeting voltage stability constraints.

[0143] The green energy-saving optimization module 6 adopts the following comprehensive objective function: ; ; in, To find the minimum value of the objective function, , , These are the weighting factors for voltage deviation, energy consumption, and lifespan loss, respectively. For voltage deviation indicators, This refers to the total active power loss of the system, including voltage regulator losses and SVG losses. This is the equipment lifespan loss factor. Lifetime sensitivity coefficient This is the exponential saturation lifetime penalty function.

[0144] The above formula mitigates the impact when lifespan loss is low, while rapidly amplifying the penalty when lifespan approaches the threshold, significantly improving the long-term operational reliability of the equipment. This formula, through a combination of quadratic, logarithmic, and exponential functions, achieves: prioritizing voltage stability without excessively sacrificing energy efficiency; ensuring energy consumption control exhibits diminishing marginal returns; and "early detection and flexible suppression" of lifespan loss. This supports the green energy-saving optimization module 6 in making reasonable online corrections to the virtual inertia coefficient, damping coefficient, and active / reactive power allocation ratio.

[0145] The core update formula of the heuristic optimization algorithm: In this embodiment, the heuristic optimization algorithm employs an improved particle swarm optimization (PSO) structure and introduces a nonlinear modulation function to enhance search stability and creativity.

[0146] Speed ​​update formula: ; ; in, This represents the velocity vector of the i-th candidate solution in the n-th iteration. This is the i-th candidate parameter vector in the n-th iteration; Let be the individual best solution obtained historically for the i-th candidate solution. This is the globally optimal solution in the current population. and All are learning factors, with values ​​ranging from (0,2); This is a sigmoid mapping function used to restrict abrupt changes in step size. This is the convergence factor related to the iteration.

[0147] Position update formula: In this formula, the hyperbolic tangent function is used to: suppress large jumps in parameters in the early stages of iteration; ensure that the parameter adjustment process is continuous and controllable; and effectively avoid system instability caused by frequent and drastic changes in network parameters.

[0148] The complete working steps of a heuristic optimization algorithm: Step S1: Initialization; Randomly generate N sets of initial parameter vectors Initialize the velocity vector ; Step S2: Objective function evaluation; For each set of parameters, calculate the corresponding Then, substituting this into the objective function f, we obtain the fitness value.

[0149] Step S3: Individual and global optimal updates; like Then update From all The global optimal solution corresponding to the minimum value of the objective function is selected.

[0150] Step S4: Iterative parameter update; Calculated according to the above velocity and position update formula Boundary trimming is performed on parameters that exceed physical constraints.

[0151] Step S5: Output and Feedback; When the preset number of iterations or convergence conditions are reached: output the optimal parameter correction value; and update the network control module 2 and the cross-device power distribution module 4 through the output and feedback unit 63 respectively.

[0152] Technical benefits of heuristic optimization algorithms: By introducing the above-mentioned heuristic optimization algorithm in Example 4, the following key technical effects were achieved: 1. "On-demand adjustment" of virtual inertia and damping: avoid oscillations caused by too small inertia under weak power grids, or slow response caused by too large inertia under strong power grids.

[0153] 2. Energy efficiency self-optimization of active / reactive power allocation: Under the premise of meeting voltage regulation constraints, prioritize reducing system losses.

[0154] 3. Proactive protection of equipment lifespan: By optimizing lifespan indicators, the frequent tapping of the voltage regulator 3 and the accumulation of thermal stress in the SVG are suppressed.

[0155] 4. Enhance the overall creativity and non-obviousness of the system: It organically integrates multi-source state perception, multi-objective optimization and cross-device collaborative control, which is significantly different from traditional single-objective or static tuning schemes.

[0156] (iv) Technical effects of green energy-saving optimization module 6; By introducing the green energy-saving optimization module 6 and implementing the above-mentioned working mechanism, this embodiment can achieve the following: significantly reduce the overall energy consumption of the system while ensuring that the voltage stability index does not deteriorate; reduce the frequent tapping of the voltage regulator 3 and reduce mechanical wear; reduce the thermal stress caused by the long-term high-load operation of the SVG; and enable the system to have the energy-saving operation capability of "self-learning and self-adjustment".

[0157] III. Structure and working principle of fault collaborative processing module 7; (a) Module structure; Fault collaborative processing module 7 includes: The fault detection and identification unit 71 is used to monitor power grid parameters in real time and generate a fault trigger signal when the voltage drop exceeds the first amplitude threshold. In this embodiment, the first amplitude threshold can be set to 10% of the rated voltage.

[0158] The multi-stage instruction generation unit 72 is used to generate and output multi-stage cooperative control instructions according to a preset timing strategy.

[0159] (ii) Multi-stage fault collaborative control logic; Upon detecting a voltage drop event, the system executes control in the following stages: First stage (rapid support stage): Send the first control command to the static var generator 5 to make it prioritize outputting rated reactive power not lower than the first proportional threshold. In this embodiment, the first proportional threshold is set to 60% to 80% of the rated reactive capacity; At the same time, a second control command is sent to voltage regulator 3, which uses the VSG algorithm to maintain the virtual power angle stability and suppress active power oscillation.

[0160] Second stage (power balance stage): Send a third control command to the static var generator 5 so that it absorbs active power at a rated capacity not exceeding the second proportional threshold, based on the redundant active power calculated by the cross-device power distribution module 4. In this embodiment, the second proportional threshold is set to 30% of the SVG's rated capacity.

[0161] Fault recovery phase (smooth switching phase): When the voltage is detected to have recovered to above the second amplitude threshold (e.g., 95% of the rated voltage). Send a fourth control command to the static var generator 5 to control it to release the absorbed active power at a rate not greater than a preset rate. The fifth control command is sent synchronously to the voltage regulator 3 to control it to gradually adjust the tap position and restore it to the optimal steady-state operating point.

[0162] During the fault recovery phase, the SVG active power release rate adopts an exponential decay + linear limiting function, as shown in the formula: ; ; in, The active power released at time t specifically means the active power released by the SVG to the system at time t during the fault recovery phase. This refers to the active power release attenuation coefficient, specifically a time-decrease parameter that controls the rate of active power release. The larger the size, the faster the release. The smaller the size, the more gradual the release; It is an exponential decay function, used to ensure that the active power release is continuous and monotonically decreasing, while avoiding voltage or frequency oscillations caused by sudden changes; This is the active power release time limit parameter, specifically meaning the minimum time scale for SVG to safely release active power. It is used to limit the early release rate and prevent excessively rapid feedback. For piecewise limiting functions, in At that time, the release range is limited; At that time, the time constraint is lifted.

[0163] By setting the above formula, power backlash during fault recovery can be prevented, effectively achieving a smooth, shock-free switching.

[0164] (III) Technical effects of fault collaborative processing module 7; Through the above-described multi-stage collaborative processing mechanism, this embodiment can achieve the following: Rapidly establish reactive power support in the early stages of voltage drop to suppress voltage collapse; During the fault, the active power regulation pressure should be reasonably distributed to avoid excessive operation of the voltage regulator 3; Achieve a smooth, shock-free, and oscillation-free transition during fault recovery; In conjunction with the green energy-saving optimization module 6, it avoids frequent fault triggering that could lead to long-term deterioration of energy consumption and lifespan indicators.

[0165] IV. Overall technical effects of Example 4; By combining the synergistic effects of the green energy-saving optimization module 6 and the fault collaborative processing module 7, this embodiment enables the voltage regulator 3 and the SVG collaborative voltage stabilization system to not only possess high dynamic stability, but also: The invention offers energy-saving optimization capabilities across the entire lifecycle; rapid adaptive capabilities for extreme operating conditions; and the advantage of coordinated control with decoupled parameter and execution layers but unified target layers. This implementation further highlights the substantial progress made by the present invention in system-level intelligent coordination and green operation compared to traditional voltage stabilization solutions in weak grid photovoltaic grid-connected scenarios.

[0166] The above are merely preferred embodiments of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principle of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A voltage regulator and SVG coordinated voltage stabilization system for photovoltaic power plants, characterized in that, include: The power grid strength detection module is used to collect power grid parameters in real time and calculate the real-time short-circuit ratio to determine the power grid strength level. A grid-type control module, connected to the power grid strength detection module, is used to generate a virtual inertia coefficient and a damping coefficient based on the real-time short-circuit ratio using a virtual inertia adaptive formula. A voltage regulator, connected to the grid-type control module, is used to perform dynamic tap adjustment in a grid-type mode according to the virtual inertia coefficient and the damping coefficient. The cross-device power distribution module is connected to the power grid strength detection module and the grid-type control module. It is used to obtain the real-time load rate of the voltage regulator and the real-time load rate of the static var generator. Based on the power grid strength level, the real-time short-circuit ratio, the real-time load rate of the voltage regulator and the real-time load rate of the static var generator, the active power distribution ratio and the reactive power distribution ratio are calculated. A static var generator is connected to the cross-device power distribution module and is used to switch between reactive power compensation and active power absorption / release according to the active power distribution ratio and the reactive power distribution ratio, and to undertake the corresponding reactive power regulation task. The green energy-saving optimization module connects the network control module and the cross-device power distribution module. It is used to dynamically optimize and correct the virtual inertia coefficient, damping coefficient, active power distribution ratio and reactive power distribution ratio with system voltage deviation, total loss and equipment life loss as optimization targets, and then feeds the results back to the corresponding modules. The fault coordination processing module is connected to the power grid strength detection module, the voltage regulator and the static var generator. When the power grid voltage in the power grid parameters is detected to drop above a preset threshold, the module coordinates and controls the static var generator and the voltage regulator to perform a fault ride-through action. When the voltage recovers, the module controls the two to coordinate their actions to switch to steady-state operation without impact.

2. The voltage regulator and SVG coordinated voltage stabilization system for photovoltaic power plants according to claim 1, characterized in that, The power grid strength detection module includes: The first acquisition unit is used to synchronously acquire the three-phase voltage, three-phase current and frequency signals at the grid connection point at a frequency not lower than the first preset frequency. An impedance calculation unit, connected to the first acquisition unit, is used to estimate the equivalent impedance of the power grid online based on the recursive least squares method according to the three-phase voltage and three-phase current. The short-circuit ratio calculation unit, connected to the impedance calculation unit, is used to calculate the real-time short-circuit ratio based on the equivalent impedance and the rated capacity of the photovoltaic power station. The strength determination unit, connected to the short-circuit ratio calculation unit, is used to compare the real-time short-circuit ratio with a first strength threshold to determine the power grid strength level; wherein, when the real-time short-circuit ratio is greater than the first strength threshold, it is determined to be a strong power grid, otherwise it is determined to be a weak power grid. The data verification and buffering unit is connected to the first acquisition unit and the impedance calculation unit respectively. It is used to verify the validity of the acquired three-phase voltage and three-phase current, and to buffer the verified data using a sliding time window to form a time-series data sequence for calculation. The impedance calculation unit is further used to perform online estimation of the equivalent impedance based on the time-series data sequence by using a recursive least squares method with a forgetting factor.

3. The voltage regulator and SVG coordinated voltage stabilization system for photovoltaic power plants according to claim 1, characterized in that, The network-type control module includes: A parameter receiving unit is used to receive the real-time short-circuit ratio from the power grid strength detection module; A virtual inertia calculation unit, connected to the parameter receiving unit, is used to calculate the virtual inertia coefficient based on the real-time short-circuit ratio and a preset adaptive formula, and to perform boundary constraint processing on the virtual inertia coefficient to output the final virtual inertia coefficient; wherein, the adaptive formula is configured such that when the real-time short-circuit ratio is less than or equal to a first short-circuit threshold, the virtual inertia coefficient increases as the real-time short-circuit ratio decreases; The damping coefficient calculation unit is connected to the parameter receiving unit and the virtual inertia calculation unit, respectively, and is used to calculate the damping coefficient that matches the current power grid strength level and virtual inertia based on the virtual inertia coefficient and the real-time short-circuit ratio through a dynamic matching algorithm. The output and update unit is connected to the virtual inertia calculation unit and the damping coefficient calculation unit, respectively. It is used to send the virtual inertia coefficient and the damping coefficient to the voltage regulator and receive correction instructions from the green energy-saving optimization module. It updates the virtual inertia coefficient and the damping coefficient online according to the correction instructions.

4. The voltage regulator and SVG coordinated voltage stabilization system for photovoltaic power plants according to claim 1, characterized in that, The voltage regulator includes: A control interface unit is used to receive the virtual inertia coefficient and the damping coefficient from the network-type control module; The VSG algorithm execution unit is connected to the control interface unit. It stores a virtual synchronous generator control algorithm, which is used to simulate the rotor motion equation of the synchronous generator according to the virtual inertia coefficient and the damping coefficient, and calculate and output the virtual power angle and virtual frequency to characterize the grid synchronization state in real time. The tap control unit, connected to the VSG algorithm execution unit, is used to generate tap adjustment commands based on the virtual power angle and the virtual frequency, combined with the grid connection point voltage feedback signal, and dynamically adjust the action rate of executing the tap adjustment commands according to the virtual inertia coefficient. The collaborative interface unit, connected to the VSG algorithm execution unit, is used to receive fault crossing instructions under the coordination of the fault collaborative processing module, and adjust the control target of the VSG algorithm execution unit according to the fault crossing instructions, so as to prioritize maintaining the stability of the virtual power angle during the fault.

5. The voltage regulator and SVG coordinated voltage stabilization system for photovoltaic power plants according to claim 1, characterized in that, The cross-device power allocation module includes: The data fusion unit is used to receive the grid strength level and the real-time short-circuit ratio output by the grid strength detection module, and to obtain the real-time load rate of the voltage regulator and the real-time load rate of the static var generator. The allocation ratio calculation unit, connected to the data fusion unit, is used to take the real-time short-circuit ratio, the real-time load rate of the voltage regulator, and the real-time load rate of the static var generator as input variables, calculate them through a preset collaborative allocation algorithm, and output the active power allocation ratio and the reactive power allocation ratio; wherein, the collaborative allocation algorithm is configured to dynamically determine the active power allocation ratio based on the degree to which the real-time short-circuit ratio deviates from the reference value and the difference between the two load rates. The collaborative allocation algorithm includes an active power allocation strategy, which is configured as follows: When the calculated active power allocation ratio is greater than zero, the active power allocation ratio is used to indicate that the static var generator absorbs active power, and the proportion of absorbed power does not exceed a first proportion threshold; when the active power allocation ratio is less than or equal to zero, the active power allocation ratio is used to indicate that the static var generator does not undertake the task of absorbing active power.

6. The voltage regulator and SVG coordinated voltage stabilization system for photovoltaic power plants according to claim 5, characterized in that, The process by which the allocation ratio calculation unit executes the collaborative allocation algorithm specifically includes: The real-time short-circuit ratio is processed based on the first weighting coefficient to obtain the first component characterizing the influence of power grid strength; The difference between the real-time load rate of the voltage regulator and the real-time load rate of the static var generator is processed based on the second weighting coefficient to obtain a second component characterizing the impact of equipment load balance. The first component and the second component are combined to obtain the active power allocation ratio; and the reactive power allocation ratio is determined based on the active power allocation ratio.

7. The voltage regulator and SVG coordinated voltage stabilization system for photovoltaic power plants according to claim 1, characterized in that, The static var generator includes: A control command receiving unit is used to receive the active power allocation ratio and the reactive power allocation ratio from the cross-device power allocation module; The four-quadrant mode control unit is connected to the control command receiving unit and is used to control the static var generator to switch between reactive power compensation mode, active power absorption mode and active power release mode according to the active power distribution ratio, and to determine the corresponding power command. The power execution unit is connected to the four-quadrant mode control unit and is used to execute the power command and output the corresponding reactive power and active power to undertake the reactive power adjustment task determined by the reactive power allocation ratio. The fault coordination unit, connected to the power execution unit, is used to, upon receiving a priority support instruction from the fault coordination processing module, exceed the conventional constraints of the active power allocation ratio and the reactive power allocation ratio, and prioritize controlling the power execution unit to output a rated reactive power capacity not lower than a second ratio threshold, so as to quickly support the grid voltage. The four-quadrant mode control unit is specifically used for: When the active power allocation ratio is greater than zero, the static var generator is controlled to enter a mixed operation mode of active power absorption and reactive power compensation, wherein the upper limit of the absorbed active power does not exceed the product of the active power allocation ratio and a preset capacity reference value. When the active power allocation ratio is less than or equal to zero, the static var generator is controlled to enter the pure reactive power compensation mode.

8. The voltage regulator and SVG coordinated voltage stabilization system for photovoltaic power plants according to claim 1, characterized in that, The green energy-saving optimization module includes: The second acquisition unit is used to acquire real-time operating parameters of the system. The real-time operating parameters of the system include at least: deviation indicators characterizing voltage stability, loss indicators characterizing system efficiency, and life indicators characterizing equipment operating status. A multi-objective optimization unit, connected to the second acquisition unit, is used to perform online iterative optimization calculations on the virtual inertia coefficient and damping coefficient from the network control module, and the active power allocation ratio and reactive power allocation ratio from the cross-device power allocation module, based on a preset heuristic optimization algorithm, using the deviation index, the loss index and the lifetime index as optimization objectives, and generating a set of optimized parameter correction values. The output and feedback unit, connected to the multi-objective optimization unit, is used to send the optimized parameter correction values ​​to the network control module and the cross-device power allocation module respectively, so as to update their output parameters.

9. The voltage regulator and SVG coordinated voltage stabilization system for photovoltaic power plants according to claim 8, characterized in that, The optimization objective configured in the multi-objective optimization unit is to simultaneously minimize the weighted sum of the deviation index, the loss index, and the lifetime index. The heuristic optimization algorithm is configured to perform parallel search of the solution space of the virtual inertia coefficient, the damping coefficient, the active power allocation ratio, and the reactive power allocation ratio within a preset iteration period, in order to find the parameter correction value that satisfies the optimization objective. The deviation index is the deviation between the grid connection point voltage and the rated voltage; the loss index is the total active power loss of the system, including the voltage regulator loss and the static var generator loss; and the lifespan index is a comprehensive loss factor calculated based on the number of times the voltage regulator operates and the thermal load state of the static var generator.

10. The voltage regulator and SVG coordinated voltage stabilization system for photovoltaic power plants according to claim 1, characterized in that, The fault collaborative processing module includes: The fault detection and identification unit is used to determine in real time whether a voltage drop event has occurred based on the power grid parameters, and to generate a fault trigger signal when the voltage drop exceeds a first amplitude threshold. A multi-stage instruction generation unit, connected to the fault detection and identification unit, is used to generate and output multi-stage collaborative control instructions in response to the fault trigger signal, according to a preset timing strategy. The timing strategy includes: In the first stage, a first control command is sent to the static var generator to make it prioritize outputting rated reactive power not lower than a first proportional threshold; at the same time, a second control command is sent to the voltage regulator to make it maintain virtual power angle stability. In the second stage, a third control command is sent to the static var generator to enable it to absorb active power at its rated capacity, not exceeding the second proportional threshold, based on the redundant active power calculated by the cross-device power allocation module. During the fault recovery phase, when the voltage is detected to have recovered to above the second amplitude threshold, a fourth control command is sent to the static var generator to control it to release the absorbed active power at a rate not greater than a preset rate, and a fifth control command is simultaneously sent to the voltage regulator to control it to adjust the tap position.

Images