An inverter control method and system based on multi-dimensional evaluation

Abstract

This application provides an inverter control method and system based on multi-dimensional evaluation. The method includes: acquiring the operating data sequence of a target inverter; determining the reactive power target value of the target inverter according to a preset inverter control model and the system operating parameters of the power grid; extracting the instantaneous reactive power data sequence of the target inverter during a preset time period, the steady-state reactive power baseline value before the fault, the steady-state reactive power average value during the fault, and the reactive power peak value during the fault from the operating data sequence, and then calculating several performance indicators of the target inverter under the fault scenario based on the extracted data; determining the performance evaluation level of the target inverter based on each performance indicator; adjusting the current control parameters of the target inverter according to the performance evaluation level, generating and controlling the target inverter according to the optimized control parameters of the target inverter, thereby improving the operating reliability and stability of the inverter.

Description

Technical Field

[0001] This application relates to the field of power system grid connection technology, and in particular to an inverter control method and system based on multi-dimensional evaluation. Background Technology

[0002] With the accelerated integration of high-proportion renewable energy into the grid, the power system faces severe challenges such as declining inertia and weakened disturbance resistance. Traditional grid-following inverters, due to their passive following characteristics, can no longer meet the requirements for stable operation, making grid-connected photovoltaic inverters with autonomous synchronization and active support capabilities a key solution. The core of this technology lies in using virtual synchronous generator (VSG) control technology to flexibly adjust active and reactive power, simulating the inertia and damping characteristics of a synchronous generator to support grid voltage and frequency. However, while this technology provides the equipment with strong flexibility, it also introduces complex technical challenges: the natural fluctuations in photovoltaic output and the grid frequency disturbances responding to provide inertia support can easily lead to DC-side power imbalance and voltage instability; mismatch between VSG control parameters (such as virtual inertia and damping coefficient) and the real-time changing grid impedance characteristics can easily trigger system oscillations; especially under fault conditions such as low-voltage ride-through, the dynamic coordination between forced reactive current injection on the grid side and DC-side voltage stability is a significant technical challenge. The root of these problems lies in the fact that the superior performance of grid-connected inverters is highly dependent on the precise matching and dynamic adjustment of their control parameters, while existing fixed-parameter control strategies are difficult to adapt to complex and ever-changing actual operating conditions.

[0003] Currently, parameter tuning for grid-connected inverters mainly relies on simulation-based offline design and trial-and-error based on field experience. While the former allows for theoretical analysis, it struggles to accurately simulate the complex dynamic characteristics of the actual power grid, often resulting in poor performance of simulation-optimized parameters in field applications. The latter is inefficient and lacks systematicity, failing to guarantee global optimality under different operating points and disturbances. Although existing research has focused on the impact of VSG parameters on stability or specific performance, these methods generally lack a closed-loop, adaptive adjustment mechanism based on real-time operational feedback. Once the equipment is connected to the grid, its control parameters become relatively fixed, making online calibration and optimization impossible based on actual support effects and changes in grid conditions. This leads to the failure to fully realize its active support potential and may even trigger new stability risks under certain circumstances due to parameter incompatibility. Summary of the Invention

[0004] To address the aforementioned technical issues, this application provides an inverter control method and system based on multi-dimensional evaluation, thereby improving the operational reliability and stability of the inverter.

[0005] In a first aspect, embodiments of this application provide an inverter control method based on multi-dimensional evaluation, including: Obtain the operating data sequence of the target inverter under a preset time period in a fault scenario; Based on the preset inverter control model and the system operating parameters of the power grid where the target inverter is located, the reactive power target value of the target inverter is determined; Extract the instantaneous reactive power data sequence of the target inverter during a preset time period, the steady-state reactive power reference value before the fault, the steady-state reactive power average value during the fault, and the reactive power peak value during the fault from the operating data sequence. Based on the instantaneous reactive power data sequence, the target reactive power value, the steady-state reactive power reference value, the steady-state reactive power average value, and the reactive power peak value, several performance indicators of the target inverter under fault scenarios are calculated. The performance evaluation level of the target inverter is determined based on each of the aforementioned performance indicators; The current control parameters of the target inverter are adjusted according to the performance evaluation level, optimized control parameters of the target inverter are generated, and the target inverter is controlled according to the optimized control parameters and the inverter control model.

[0006] This application provides an inverter control method based on multi-dimensional evaluation. By acquiring operating data sequences under fault scenarios and dynamically determining the reactive power target value in conjunction with grid system parameters, it achieves multi-dimensional quantitative evaluation and closed-loop optimization control of inverter performance. The advantage of this embodiment lies in upgrading the traditional fixed-parameter control strategy to an adaptive mechanism based on real-time feedback, thereby significantly improving the inverter's operational reliability and response accuracy under complex operating conditions. Specifically, by extracting instantaneous reactive power data sequences and key features such as steady-state reference values, average values, and peak values ​​from the operating data sequences, the dynamic response capability of the inverter is comprehensively quantified. Then, performance levels are determined based on multi-dimensional performance indicators, and control parameters are dynamically adjusted according to the performance levels, realizing a shift from passive response to active optimization. This embodiment can not only effectively cope with uncertainties such as photovoltaic output fluctuations and grid impedance changes, but also optimize the inverter's control parameters in real time based on operating data under fault scenarios such as low-voltage ride-through, improving the inverter's subsequent operational reliability and stability.

[0007] Furthermore, acquiring the operating data sequence of the target inverter over a preset time period under fault scenarios includes: Acquire the initial three-phase voltage data sequence, initial three-phase current data sequence, and initial grid voltage data sequence of the target inverter under a preset time period in a fault scenario, and construct the initial data sequence. Each data sequence in the initial data sequence is downsampled to obtain a running data sequence with a preset time resolution.

[0008] This application further refines the method for acquiring operational data sequences. By downsampling, the initial high-resolution data is converted into a sequence with a preset time resolution, significantly improving the efficiency and practicality of data processing. In fault scenarios, power systems often generate massive amounts of real-time data. Directly processing the raw, high-sampling-rate data not only incurs a heavy computational burden but may also affect the accuracy of analysis due to noise interference. Downsampling effectively compresses the data size while preserving key dynamic characteristics, reducing storage and computing resource requirements, making this embodiment more suitable for practical engineering applications. Furthermore, by reasonably setting the time resolution, the data processing speed can be accelerated while ensuring analytical accuracy, thereby providing timely support for real-time control decisions. This approach not only improves the system's response efficiency but also lays the technical foundation for its widespread application in multi-inverter collaboration or large-scale power grids, improving the efficiency of subsequent multi-dimensional evaluation and parameter optimization.

[0009] In one possible implementation, determining the target reactive power value of the target inverter based on a preset inverter control model and the system operating parameters of the power grid where the target inverter is located includes: Calculate the quotient of the system rated capacity and the system rated voltage in the system operating parameters to obtain the first intermediate value; The grid voltage value at the time of the fault occurrence is obtained from the operating data sequence, and the difference between the system rated voltage and the grid voltage value is calculated to obtain a second intermediate value; The forced injection coefficient of the target inverter under the fault scenario is determined based on the inverter control model. Multiply the first intermediate value, the second intermediate value, and the forced injection coefficient to obtain the reactive power target value of the target inverter.

[0010] This application clarifies the calculation method for the reactive power target value. By combining system rated parameters, real-time grid voltage, and forced injection coefficients in the inverter control model, a scientific and dynamic determination of the target value is achieved. In traditional methods, the reactive power target value usually relies on fixed thresholds or empirical settings, which are difficult to adapt to real-time changes in grid conditions and can easily lead to insufficient support or over-response. This embodiment introduces an inverter control model. During the inverter's fault ride-through process, this model can dynamically generate the corresponding reactive power target value based on the ratio of system rated capacity to voltage, grid voltage deviation, and modeling coefficients under fault scenarios. This ensures that the generated reactive power target value accurately reflects the actual demand of the current grid, and the inverter is controlled to complete fault ride-through based on this reactive power target value. Therefore, in the evaluation process of this embodiment, the accurate reactive power target value is directly determined based on the inverter control model according to the above control process, providing accurate basic data for subsequent multi-dimensional evaluation and parameter optimization.

[0011] In one possible implementation, extracting the instantaneous reactive power data sequence of the target inverter over a preset time period, the steady-state reactive power reference value before the fault, the steady-state reactive power average value during the fault, and the reactive power peak value during the fault from the operating data sequence includes: The three-phase voltage data sequence and the three-phase current data sequence in the running data sequence are respectively subjected to coordinate transformation to obtain the voltage direct axis component data sequence, the voltage quadrature axis component data sequence, the current direct axis component data sequence, and the current quadrature axis component data sequence; Based on the voltage direct-axis component data sequence, the voltage quadrature-axis component data sequence, the current direct-axis component data sequence, and the current quadrature-axis component data sequence, the corresponding instantaneous reactive power data sequence is calculated based on instantaneous reactive power theory. According to the preset first time window, the first subsequence before the fault is extracted from the instantaneous reactive power data sequence, and the steady-state reactive power reference value is calculated based on the first subsequence. According to the preset second time window, a second subsequence during the fault period is extracted from the instantaneous reactive power data sequence, and the steady-state reactive power average value is calculated based on the second subsequence; According to the preset third time window, a third subsequence during the fault period is extracted from the instantaneous reactive power data sequence, and the maximum instantaneous reactive power in the third subsequence is determined as the reactive power peak value.

[0012] This application describes in detail the process of extracting key reactive power features from operating data, including methods for calculating instantaneous reactive power sequences, steady-state reference values, average values, and peak values. Through coordinate transformation and instantaneous reactive power theory, this embodiment can accurately separate the active and reactive components in voltage and current, thereby obtaining a true and continuous dynamic trajectory of reactive power. Based on this, by combining a preset time window to segment and statistically analyze data before and during a fault, the steady-state and transient characteristics of the inverter are further quantified. This refined feature extraction method not only provides a reliable data foundation for subsequent performance evaluation but also helps identify potential problems in the inverter's fault response process, such as response delay, overshoot, or steady-state error. Through comprehensive analysis of multi-dimensional features, this method achieves a comprehensive characterization of inverter behavior, providing a scientific basis for the formulation of subsequent optimized control strategies.

[0013] In one possible implementation, the calculation of several performance indicators of the target inverter under fault scenarios based on the instantaneous reactive power data sequence, the target reactive power value, the steady-state reactive power reference value, the steady-state reactive power average value, and the reactive power peak value includes: The reactive power response time of the target inverter after a fault occurs is determined based on the steady-state reactive power reference value, the reactive power data sequence, and the reactive power target value. The reactive power support coefficient and steady-state error of the target inverter are calculated based on the steady-state average reactive power value and the target reactive power value. The overshoot of the target inverter is determined based on the peak reactive power and the target reactive power value. Starting from the fault initiation time, the instantaneous reactive power data sequence is traversed, and the first adjustment time that meets the preset conditions is determined. The adjustment time of the target inverter is determined based on the difference between the adjustment time and the fault initiation time. The reactive power response time, the reactive power support coefficient, the steady-state error, the overshoot, and the adjustment time are used as several performance indicators of the target inverter under fault scenarios.

[0014] This application provides methods for calculating several key performance indicators of inverters under fault scenarios, including reactive power response time, reactive power support coefficient, steady-state error, overshoot, and settling time. These indicators comprehensively reflect the dynamic response capability and steady-state performance of the inverter from different dimensions, providing specific and operable standards for quantitative evaluation. For example, reactive power response time measures the inverter's rapid response capability to faults; reactive power support coefficient and steady-state error reflect its steady-state support accuracy; and overshoot and settling time characterize the stability and convergence speed during transient processes. Through comprehensive evaluation of multiple indicators, this embodiment can identify the performance shortcomings of the inverter under fault scenarios, further improving the operational reliability and stability of the inverter.

[0015] Furthermore, the preset condition is that within a preset duration after the adjustment time, the instantaneous values ​​of each reactive power in the reactive power data sequence are always within a preset error band.

[0016] This application refines the conditions for determining the adjustment time in the performance indicators, stipulating that the instantaneous value of reactive power must remain within a preset error band for a period of time after the adjustment time for the adjustment to be considered complete. This condition strictly defines the standard for the system to enter steady state, avoiding misjudgments caused by instantaneous fluctuations or noise, thereby ensuring the accuracy and reliability of the adjustment time calculation. By introducing dual constraints of error band and duration, this method can effectively distinguish between true steady-state convergence and temporary fluctuations, improving the robustness of the evaluation results. By precisely quantifying the adjustment time, control parameters can be further optimized, improving the stability and efficiency of the inverter during fault recovery.

[0017] Furthermore, determining the performance evaluation level of the target inverter based on each of the performance indicators includes: The performance score corresponding to each performance indicator is calculated based on the piecewise function corresponding to each performance indicator. The performance scores are weighted and summed according to preset weighting coefficients to obtain the comprehensive performance score of the target inverter. The performance evaluation level of the target inverter is determined based on the comprehensive performance score.

[0018] This application proposes a performance evaluation level determination method based on piecewise functions and weighted summation. By transforming multi-dimensional performance indicators into a unified score and calculating a comprehensive performance score using weighted coefficients, the evaluation level is ultimately determined. This method overcomes the limitations of single-indicator evaluation, achieving a scientific fusion and overall evaluation of multiple indicators. The piecewise function design allows for differentiated scoring rules for different indicators within different value ranges, better aligning with actual engineering needs. The introduction of weighted coefficients reflects the relative importance of different indicators in the evaluation, enhancing the rationality and flexibility of the assessment. Through comprehensive scoring and level division, this method can intuitively reflect the overall performance level of the inverter and formulate differentiated optimization strategies for different levels, thereby achieving refined and hierarchical control optimization.

[0019] In one possible implementation, adjusting the current control parameters of the target inverter according to the performance evaluation level to generate optimized control parameters for the target inverter includes: If the performance evaluation level is the first level, then the current control parameters of the target inverter will be used as the optimized control parameters of the target inverter. If the performance evaluation level is the second level, then the droop coefficient in the current control parameter is increased and the response time constant in the current control parameter is shortened, thereby generating the optimized control parameters for the target inverter; If the performance evaluation level is level three, the current margin in the current control parameters is increased, the optimized control parameters of the target inverter are generated, and a collaborative support request is sent to the grid dispatch system where the target inverter is located.

[0020] This application's embodiments establish differentiated control parameter adjustment strategies based on performance evaluation levels, employing measures such as maintenance, optimization, or collaborative support for different levels. For example, for the first level, the current parameters are maintained; for the second level, the droop coefficient is increased and the response time constant is shortened; for the third level, the current margin is increased and grid collaborative support is requested. This hierarchical response mechanism achieves refined and adaptive control strategies, avoiding unnecessary parameter adjustments and enabling timely reinforcement measures when performance is insufficient. By dynamically adjusting key parameters such as the droop coefficient and response time constant, the dynamic response characteristics and steady-state accuracy of the inverter can be significantly improved; and in scenarios requiring collaborative support, the linkage with the grid dispatch system further enhances the overall stability and recovery capability of the system, thereby further improving the reliability and stability of inverter operation.

[0021] Secondly, embodiments of this application provide an inverter control system based on multi-dimensional evaluation, including an acquisition module, a target determination module, a data extraction module, an index calculation module, an evaluation module, and a control module; The acquisition module is used to acquire the operating data sequence of the target inverter during a preset time period under fault scenarios; The target determination module is used to determine the reactive power target value of the target inverter based on the preset inverter control model and the system operating parameters of the power grid where the target inverter is located; The data extraction module is used to extract from the operating data sequence the instantaneous reactive power data sequence of the target inverter during a preset time period, the steady-state reactive power reference value before the fault, the steady-state reactive power average value during the fault, and the reactive power peak value during the fault. The index calculation module is used to calculate several performance indicators of the target inverter under fault scenarios based on the instantaneous reactive power data sequence, the target reactive power value, the steady-state reactive power reference value, the steady-state reactive power average value, and the reactive power peak value. The evaluation module is used to determine the performance evaluation level of the target inverter based on each of the performance indicators; The control module is used to adjust the current control parameters of the target inverter according to the performance evaluation level, generate optimized control parameters for the target inverter, and control the target inverter according to the optimized control parameters and the inverter control model.

[0022] Furthermore, the indicator calculation module calculates several performance indicators of the target inverter under fault scenarios based on the instantaneous reactive power data sequence, the target reactive power value, the steady-state reactive power reference value, the steady-state reactive power average value, and the reactive power peak value, including: The reactive power response time of the target inverter after a fault occurs is determined based on the steady-state reactive power reference value, the reactive power data sequence, and the reactive power target value. The reactive power support coefficient and steady-state error of the target inverter are calculated based on the steady-state average reactive power value and the target reactive power value. The overshoot of the target inverter is determined based on the peak reactive power and the target reactive power value. Starting from the fault initiation time, the instantaneous reactive power data sequence is traversed, and the first adjustment time that meets the preset conditions is determined. The adjustment time of the target inverter is determined based on the difference between the adjustment time and the fault initiation time. The reactive power response time, the reactive power support coefficient, the steady-state error, the overshoot, and the adjustment time are used as several performance indicators of the target inverter under fault scenarios. Attached Figure Description

[0023] Figure 1 A flowchart illustrating an inverter control method based on multi-dimensional evaluation provided in an embodiment of this application; Figure 2 A schematic diagram of the inverter dynamic reactive power response curve in a test scenario for an inverter control method based on multi-dimensional evaluation provided in this application embodiment; Figure 3 A schematic diagram comparing inverter performance indicators in a test scenario for an inverter control method based on multi-dimensional evaluation provided in this application embodiment; Figure 4 A radar chart of the multi-dimensional performance of an inverter in a test scenario, illustrating an inverter control method based on multi-dimensional evaluation provided in this application embodiment. Figure 5 A schematic diagram of a grid voltage disturbance condition under a test scenario for an inverter control method based on multi-dimensional evaluation provided in this application embodiment: Figure 6 This is a schematic diagram of an inverter control system based on multi-dimensional evaluation, provided as an embodiment of this application. Detailed Implementation

[0024] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0025] It should be noted that the step numbers in this document are only for the convenience of explaining the specific embodiments and are not intended to limit the order in which the steps are performed. In the description of this application, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature specified as "first" or "second" may explicitly or implicitly include one or more of that feature.

[0026] Example 1: like Figure 1 As shown, Embodiment 1 provides an inverter control method based on multi-dimensional evaluation, including steps S1-S6: Step S1: Obtain the operating data sequence of the target inverter during a preset time period under fault scenarios; Step S2: Determine the target reactive power value of the target inverter based on the preset inverter control model and the system operating parameters of the power grid where the target inverter is located; Step S3: Extract the instantaneous reactive power data sequence of the target inverter during a preset time period, the steady-state reactive power reference value before the fault, the steady-state reactive power average value during the fault, and the reactive power peak value during the fault from the operating data sequence. Step S4: Based on the instantaneous reactive power data sequence, the target reactive power value, the steady-state reactive power reference value, the steady-state reactive power average value, and the reactive power peak value, calculate several performance indicators of the target inverter under fault scenarios. Step S5: Determine the performance evaluation level of the target inverter based on each of the aforementioned performance indicators; Step S6: Adjust the current control parameters of the target inverter according to the performance evaluation level, generate the optimized control parameters of the target inverter, and control the target inverter according to the optimized control parameters and the inverter control model.

[0027] This application provides an inverter control method based on multi-dimensional evaluation. By acquiring operating data sequences under fault scenarios and dynamically determining the reactive power target value in conjunction with grid system parameters, it achieves multi-dimensional quantitative evaluation and closed-loop optimization control of inverter performance. The advantage of this embodiment lies in upgrading the traditional fixed-parameter control strategy to an adaptive mechanism based on real-time feedback, thereby significantly improving the inverter's operational reliability and response accuracy under complex operating conditions. Specifically, by extracting instantaneous reactive power data sequences and key features such as steady-state reference values, average values, and peak values ​​from the operating data sequences, the dynamic response capability of the inverter is comprehensively quantified. Then, performance levels are determined based on multi-dimensional performance indicators, and control parameters are dynamically adjusted according to the performance levels, realizing a shift from passive response to active optimization. This embodiment can not only effectively cope with uncertainties such as photovoltaic output fluctuations and grid impedance changes, but also optimize the inverter's control parameters in real time based on operating data under fault scenarios such as low-voltage ride-through, improving the inverter's subsequent operational reliability and stability.

[0028] Furthermore, in step S1, obtaining the operating data sequence of the target inverter for a preset time period under a fault scenario includes: Acquire the initial three-phase voltage data sequence, initial three-phase current data sequence, and initial grid voltage data sequence of the target inverter under a preset time period in a fault scenario, and construct the initial data sequence. Each data sequence in the initial data sequence is downsampled to obtain a running data sequence with a preset time resolution.

[0029] This application further refines the method for acquiring operational data sequences. By downsampling, the initial high-resolution data is converted into a sequence with a preset time resolution, significantly improving the efficiency and practicality of data processing. In fault scenarios, power systems often generate massive amounts of real-time data. Directly processing the raw, high-sampling-rate data not only incurs a heavy computational burden but may also affect the accuracy of analysis due to noise interference. Downsampling effectively compresses the data size while preserving key dynamic characteristics, reducing storage and computing resource requirements, making this embodiment more suitable for practical engineering applications. Furthermore, by reasonably setting the time resolution, the data processing speed can be accelerated while ensuring analytical accuracy, thereby providing timely support for real-time control decisions. This approach not only improves the system's response efficiency but also lays the technical foundation for its widespread application in multi-inverter collaboration or large-scale power grids, improving the efficiency of subsequent multi-dimensional evaluation and parameter optimization.

[0030] In a preferred embodiment, the operating data sequence of the target inverter under fault scenarios can be passively acquired during actual operation, or fault scenarios can be actively simulated in a test environment to collect the operating data sequence of the target inverter under fault scenarios. Specifically, a test platform is set up in an actual engineering site or laboratory environment. The grid-connected photovoltaic inverter under test (i.e., the target inverter) should be in rated operating condition, and a grid voltage drop fault is simulated by a programmable AC power supply or a grid disturbance generator. The test conditions strictly follow the GB / T 19964-2024 standard, and the voltage drop depth is set, for example, a voltage drop of 20% or 30%, with the drop initiation time being... The duration is 625ms. A high-speed data acquisition system is used to synchronously record the three-phase voltage and three-phase current on the AC side of the inverter with a sampling period of 100μs, while the grid voltage is recorded as a reference signal. The total data acquisition time is no less than 2s to ensure coverage of the steady state before the fault, the transient state after the fault, and the entire recovery process after the fault. To reduce the impact of measurement noise, the original signal needs to be downsampled by 100 times to obtain an operating data sequence with a time resolution of 10ms.

[0031] In one possible implementation, step S2, determining the target reactive power value of the target inverter based on a preset inverter control model and the system operating parameters of the power grid where the target inverter is located, includes: Calculate the quotient of the system rated capacity and the system rated voltage in the system operating parameters to obtain the first intermediate value; The grid voltage value at the time of the fault occurrence is obtained from the operating data sequence, and the difference between the system rated voltage and the grid voltage value is calculated to obtain a second intermediate value; The forced injection coefficient of the target inverter under the fault scenario is determined based on the inverter control model. Multiply the first intermediate value, the second intermediate value, and the forced injection coefficient to obtain the reactive power target value of the target inverter.

[0032] This application clarifies the calculation method for the reactive power target value. By combining system rated parameters, real-time grid voltage, and forced injection coefficients in the inverter control model, a scientific and dynamic determination of the target value is achieved. In traditional methods, the reactive power target value usually relies on fixed thresholds or empirical settings, which are difficult to adapt to real-time changes in grid conditions and can easily lead to insufficient support or over-response. This embodiment introduces an inverter control model. During the inverter's fault ride-through process, this model can dynamically generate the corresponding reactive power target value based on the ratio of system rated capacity to voltage, grid voltage deviation, and modeling coefficients under fault scenarios. This ensures that the generated reactive power target value accurately reflects the actual demand of the current grid, and the inverter is controlled to complete fault ride-through based on this reactive power target value. Therefore, in the evaluation process of this embodiment, the accurate reactive power target value is directly determined based on the inverter control model according to the above control process, providing accurate basic data for subsequent multi-dimensional evaluation and parameter optimization.

[0033] In a preferred embodiment, the inverter control model is an improved grid-connected dual-mode switching control model. This model is built based on the voltage-reactive power droop control characteristics of the grid-connected inverter, and the specific control strategy is as follows: When the grid voltage Under normal operating conditions, using droop control mode, the target reactive power of the system is: When the grid voltage When the fault condition is met, the national standard forced injection mode is adopted, and the target reactive power of the system is as follows: in, The internationally mandated injection coefficient is set to 2.0 in this invention. The droop coefficient is 30 in this invention; This is the system's rated voltage; This is the system's rated capacity; This represents the measured value of the grid voltage received by the inverter through the voltage sensor.

[0034] Specifically, the system reactive power target value serves as a bridge connecting national standard requirements, control implementation, and performance evaluation, and is the core benchmark quantity of the multi-dimensional evaluation system in this embodiment. Its core role is reflected in the fact that all five core performance indicators mentioned in this embodiment are based on... Quantitative calculations are performed for reference; in the improved VSG dual-mode control strategy, It forms the computational basis for generating reactive power commands under fault conditions; in the comprehensive evaluation model, The normalization process is used for various indicators, including dimensionless indicators such as support coefficient, steady-state error, and overshoot. Per-unit calculations are performed on the denominator to make the evaluation results of inverters with different capacities and voltage levels comparable, thus meeting the requirements of standardized evaluation.

[0035] Furthermore, considering the current saturation limitation of grid-connected photovoltaic inverters under fault conditions, the reactive current command must be limited to within 1.2 times the rated current to ensure that the inverter operates within a safe current range while meeting national standard reactive power support requirements, achieving the dual goals of equipment safety and grid support during fault ride-through. Therefore, the following constraints are added: in, The target value of three-phase reactive current is obtained by converting the target value of reactive power; The maximum allowable current represents the maximum current limit that the inverter's power devices can continuously withstand; when Exceed At that time, the control system presses After inverse calculation restrictions This achieves a dynamic balance between reactive power support capacity and equipment safety.

[0036] Furthermore, considering the measurement delay and the first-order inertial element, the actual output reactive power is: Where s is the complex frequency variable in the Laplace transform; The response time constant is set to 12ms. To control the delay time, a value of 10ms is used.

[0037] The measured voltage includes measurement noise and harmonic interference, and its expression is: in, It refers to the theoretical fundamental effective value of the grid voltage at the inverter's grid connection point, and is the ideal input quantity in the voltage measurement model; To measure noise, fluctuations are observed in... Within the range; This is harmonic interference, which is related to the depth of the system fault.

[0038] In one possible implementation, step S3, extracting the instantaneous reactive power data sequence of the target inverter over a preset time period, the steady-state reactive power reference value before the fault, the steady-state reactive power average value during the fault, and the reactive power peak value during the fault from the operating data sequence, includes: The three-phase voltage data sequence and the three-phase current data sequence in the running data sequence are respectively subjected to coordinate transformation to obtain the voltage direct axis component data sequence, the voltage quadrature axis component data sequence, the current direct axis component data sequence, and the current quadrature axis component data sequence; Based on the voltage direct-axis component data sequence, the voltage quadrature-axis component data sequence, the current direct-axis component data sequence, and the current quadrature-axis component data sequence, the corresponding instantaneous reactive power data sequence is calculated based on instantaneous reactive power theory. According to the preset first time window, the first subsequence before the fault is extracted from the instantaneous reactive power data sequence, and the steady-state reactive power reference value is calculated based on the first subsequence. According to the preset second time window, a second subsequence during the fault period is extracted from the instantaneous reactive power data sequence, and the steady-state reactive power average value is calculated based on the second subsequence; According to the preset third time window, a third subsequence during the fault period is extracted from the instantaneous reactive power data sequence, and the maximum instantaneous reactive power in the third subsequence is determined as the reactive power peak value.

[0039] This application describes in detail the process of extracting key reactive power features from operating data, including methods for calculating instantaneous reactive power sequences, steady-state reference values, average values, and peak values. Through coordinate transformation and instantaneous reactive power theory, this embodiment can accurately separate the active and reactive components in voltage and current, thereby obtaining a true and continuous dynamic trajectory of reactive power. Based on this, by combining a preset time window to segment and statistically analyze data before and during a fault, the steady-state and transient characteristics of the inverter are further quantified. This refined feature extraction method not only provides a reliable data foundation for subsequent performance evaluation but also helps identify potential problems in the inverter's fault response process, such as response delay, overshoot, or steady-state error. Through comprehensive analysis of multi-dimensional features, this method achieves a comprehensive characterization of inverter behavior, providing a scientific basis for the formulation of subsequent optimized control strategies.

[0040] In a preferred embodiment, the acquired three-phase voltage and current signals are transformed to obtain the dq-axis components, and the instantaneous active and reactive power are calculated. A low-pass filter is used to filter out switching frequency harmonics, retaining the fundamental component. Based on the fault initiation time... The average reactive power value within the time window from 150ms to 50ms before the fault is extracted as the steady-state reference value. The average reactive power value within the time window from 50ms after the fault starts to 50ms before the fault ends is extracted as the steady-state reactive power value during the fault. Simultaneously record the peak reactive power during this period. Used for overshoot calculation.

[0041] In one possible implementation, in step S4, the calculation of several performance indicators of the target inverter under fault scenarios based on the instantaneous reactive power data sequence, the target reactive power value, the steady-state reactive power reference value, the steady-state reactive power average value, and the reactive power peak value includes: The reactive power response time of the target inverter after a fault occurs is determined based on the steady-state reactive power reference value, the reactive power data sequence, and the reactive power target value. The reactive power support coefficient and steady-state error of the target inverter are calculated based on the steady-state average reactive power value and the target reactive power value. The overshoot of the target inverter is determined based on the peak reactive power and the target reactive power value. Starting from the fault initiation time, the instantaneous reactive power data sequence is traversed, and the first adjustment time that meets the preset conditions is determined. The adjustment time of the target inverter is determined based on the difference between the adjustment time and the fault initiation time. The reactive power response time, the reactive power support coefficient, the steady-state error, the overshoot, and the adjustment time are used as several performance indicators of the target inverter under fault scenarios.

[0042] This application provides methods for calculating several key performance indicators of inverters under fault scenarios, including reactive power response time, reactive power support coefficient, steady-state error, overshoot, and settling time. These indicators comprehensively reflect the dynamic response capability and steady-state performance of the inverter from different dimensions, providing specific and operable standards for quantitative evaluation. For example, reactive power response time measures the inverter's rapid response capability to faults; reactive power support coefficient and steady-state error reflect its steady-state support accuracy; and overshoot and settling time characterize the stability and convergence speed during transient processes. Through comprehensive evaluation of multiple indicators, this embodiment can identify the performance shortcomings of the inverter under fault scenarios, further improving the operational reliability and stability of the inverter.

[0043] Furthermore, the preset condition is that within a preset duration after the adjustment time, the instantaneous values ​​of each reactive power in the reactive power data sequence are always within a preset error band.

[0044] This application refines the conditions for determining the adjustment time in the performance indicators, stipulating that the instantaneous value of reactive power must remain within a preset error band for a period of time after the adjustment time for the adjustment to be considered complete. This condition strictly defines the standard for the system to enter steady state, avoiding misjudgments caused by instantaneous fluctuations or noise, thereby ensuring the accuracy and reliability of the adjustment time calculation. By introducing dual constraints of error band and duration, this method can effectively distinguish between true steady-state convergence and temporary fluctuations, improving the robustness of the evaluation results. By precisely quantifying the adjustment time, control parameters can be further optimized, improving the stability and efficiency of the inverter during fault recovery.

[0045] In a preferred embodiment, a dynamic reactive power support capability evaluation system is constructed from five dimensions: 1) Response speed: based on reactive response time The core indicator is defined as the time required from the occurrence of voltage disturbance to the system reactive power output reaching 90% of the target value; 2) Support strength: based on the reactive support coefficient The core indicator is defined as the ratio of the actual average reactive power output during a fault to the theoretical target value. 3) Control accuracy: based on steady-state error The core indicator is defined as the relative error between the reactive power output and the target value during the steady-state phase of a fault. 4) Stability (Dynamic Quality): Based on overshoot The core indicator is defined as the percentage by which the peak reactive power exceeds the target value during the response process; 5) Adjust performance (stabilize performance): by adjusting time The core indicator is defined as the time required to enter the target reactive power value ±5% error range (and this state lasts for at least 100ms) and no longer exceed the required time.

[0046] The calculation methods for each performance indicator are as follows: 1) Reactive response time: in, This represents the instantaneous value of the reactive power output by the inverter. The mean steady-state reactive power before the fault; The target reactive power value during the fault period; The time when the fault occurred.

[0047] 2) Reactive power support coefficient: in, This represents the average actual reactive power output during the fault period; This indicates the time 50ms after the fault occurred; This indicates the time 50ms before the fault ends; The duration of the fault.

[0048] 3) Steady-state error: 4) Overshoot: in, The peak value represents the reactive power. The time domain interval indicates that the measurement window is within 200ms after the fault begins. This period covers the critical dynamic response stage in the early stage of the fault, ensuring the accuracy of the evaluation while avoiding interference from steady-state data in the middle and later stages of the fault on the peak value detection.

[0049] 5) Adjust the time: This formula indicates that after a fault occurs, the first moment occurs. This ensures that the reactive power remains constant for the next 100 ms starting from that moment. The adjustment time is the time to maintain the value within ±5% of the target value and the time difference between this moment and the moment the fault begins.

[0050] Furthermore, in step S5, determining the performance evaluation level of the target inverter based on each of the performance indicators includes: The performance score corresponding to each performance indicator is calculated based on the piecewise function corresponding to each performance indicator. The performance scores are weighted and summed according to preset weighting coefficients to obtain the comprehensive performance score of the target inverter. The performance evaluation level of the target inverter is determined based on the comprehensive performance score.

[0051] This application proposes a performance evaluation level determination method based on piecewise functions and weighted summation. By transforming multi-dimensional performance indicators into a unified score and calculating a comprehensive performance score using weighted coefficients, the evaluation level is ultimately determined. This method overcomes the limitations of single-indicator evaluation, achieving a scientific fusion and overall evaluation of multiple indicators. The piecewise function design allows for differentiated scoring rules for different indicators within different value ranges, better aligning with actual engineering needs. The introduction of weighted coefficients reflects the relative importance of different indicators in the evaluation, enhancing the rationality and flexibility of the assessment. Through comprehensive scoring and level division, this method can intuitively reflect the overall performance level of the inverter and formulate differentiated optimization strategies for different levels, thereby achieving refined and hierarchical control optimization.

[0052] In a preferred embodiment, a direct weighted comprehensive evaluation method is adopted, where the weights of each indicator are allocated according to their importance in the actual engineering execution: 1) Response speed weight The score for this dimension is calculated as follows: ; 2) Support strength weight The score for this dimension is calculated as follows: ; 3) Control precision weight The score for this dimension is calculated as follows: ; 4) Stability weight The score for this dimension is calculated as follows: ; 5) Adjust performance weights The score for this dimension is calculated as follows: ; The evaluation thresholds for each indicator are shown in the table below. Scores for each dimension are calculated using a piecewise linear scoring function. The overall score is calculated using the following formula: The grades are determined based on the total score, and the evaluation criteria are as follows: A (Excellent) B (Good) C (Pass) :D (Basically qualified) E (Unqualified) In one possible implementation, step S6, adjusting the current control parameters of the target inverter according to the performance evaluation level to generate optimized control parameters for the target inverter, includes: If the performance evaluation level is the first level, then the current control parameters of the target inverter will be used as the optimized control parameters of the target inverter. If the performance evaluation level is the second level, then the droop coefficient in the current control parameter is increased and the response time constant in the current control parameter is shortened, thereby generating the optimized control parameters for the target inverter; If the performance evaluation level is level three, the current margin in the current control parameters is increased, the optimized control parameters of the target inverter are generated, and a collaborative support request is sent to the grid dispatch system where the target inverter is located.

[0053] This application's embodiments establish differentiated control parameter adjustment strategies based on performance evaluation levels, employing measures such as maintenance, optimization, or collaborative support for different levels. For example, for the first level, the current parameters are maintained; for the second level, the droop coefficient is increased and the response time constant is shortened; for the third level, the current margin is increased and grid collaborative support is requested. This hierarchical response mechanism achieves refined and adaptive control strategies, avoiding unnecessary parameter adjustments and enabling timely reinforcement measures when performance is insufficient. By dynamically adjusting key parameters such as the droop coefficient and response time constant, the dynamic response characteristics and steady-state accuracy of the inverter can be significantly improved; and in scenarios requiring collaborative support, the linkage with the grid dispatch system further enhances the overall stability and recovery capability of the system, thereby further improving the reliability and stability of inverter operation.

[0054] In a preferred embodiment, the control parameters of the grid-connected photovoltaic inverter are dynamically determined based on the rating results: Grade A: Maintain the current sagging coefficient Response time constant ; Grade B: Optimize the sagging coefficient to... shorten the response time constant to ; Level C and below: Switch to conservative control mode, increase current margin to 1.5 times, and report to the power grid dispatch system for collaborative support.

[0055] The optimized control parameters are sent to the inverter controller in real time, and the updated control strategy is executed in subsequent operation. Performance evaluation is continuously performed to form a closed-loop optimization of evaluation-decision-control.

[0056] In a preferred embodiment, the control effect of the present invention on the inverter is verified by performing multiple fault tests on the target inverter using a testing system, wherein the testing system includes: 1) Test signal generation module: Generates voltage disturbance signals, such as 20% and 30% voltage drop conditions in the example; 2) Data acquisition module: Acquires output voltage, current and reactive power data of photovoltaic inverter with a sampling period of 100μs, and performs 100 times downsampling processing; 3) Indicator Calculation Module: Calculates indicators for each dimension according to the performance indicator calculation method proposed in this application; 4) Comprehensive Evaluation Module: The weighted comprehensive evaluation method is used to calculate the comprehensive score and assign a grade; 5) Visualization output module: Generates dynamic response curves, performance indicator comparison bar charts, multi-dimensional performance radar charts, and comprehensive scoring dashboards; 6) Control Strategy Decision Module: Based on the rating results of the comprehensive evaluation module, dynamically determine the control parameter adjustment strategy for grid-connected photovoltaic inverters: Level A maintains the current control parameters, Level B optimizes the droop coefficient and response time constant, and Level C and below switches to conservative control mode and reports to the grid dispatch system for collaborative support; 7) Adaptive Control Execution Module: Receives control parameter adjustment instructions from the control strategy decision module, and sends the optimized control parameters to the inverter controller in real time, so that the inverter executes the updated control strategy in subsequent operation; at the same time, it triggers the system to continuously return to the data acquisition module for performance review, forming a closed-loop adaptive optimization mechanism of evaluation-decision-control.

[0057] Test Scenario 1: Voltage Drop of 20% Operating Condition Test Considering the modulation ratio and voltage margin, the rated capacity of the grid-connected photovoltaic inverter is assumed to be 100kVA, the rated effective voltage is 315V, the droop factor is 30, and the response time constant is 0.012s. A schematic diagram of the inverter's dynamic reactive power response curve in this scenario is shown below. Figure 2 As shown, simulation results indicate a response time of 20ms; a support coefficient of 95%; a steady-state error between 1% and 2%; an overshoot of less than 5%; and a settling time within 50ms. A comparison of various performance indicators is shown in the figure below. Figure 3 As shown, the multi-dimensional performance radar chart after conversion to performance scoring is as follows: Figure 4 As shown, the target inverter achieved a comprehensive score of 97.9 points, with a rating of A, indicating good performance.

[0058] Test Scenario 2: Voltage Drop of 30% Operating Condition Test Assuming the photovoltaic inverter parameters are the same as in test scenario 1, the schematic diagram of the grid voltage disturbance condition is as follows: Figure 5 As shown, the simulation results indicate that, except for the steady-state error and settling time which fluctuate slightly within permissible limits, the other indicators are close to or equal to the test results under the 20% voltage drop condition. The overall score is 94.2 points, and the rating is also A.

[0059] Based on the evaluation results of test scenario 1 (Grade A, 97.9 points), the system decision maintains the original control parameters unchanged; based on the evaluation results of test scenario 2 (Grade A, 94.2 points), the original parameters are also maintained. If an evaluation result is Grade B (e.g., a comprehensive score of 85 points and a response time of 35ms), the optimization process is automatically triggered: the droop coefficient is adjusted from 30 to 35, the response time constant is shortened from 12ms to 10ms, and the improvement effect is verified in the next voltage drop event. After optimization, the expected response time is shortened to within 25ms, and the comprehensive score is improved to above 90 points, reaching the Grade A standard, forming a closed-loop adaptive mechanism of "evaluation-optimization-verification".

[0060] In summary, compared with the prior art, the beneficial effects of the embodiments of this application are reflected in the following four aspects: First, a systematic five-dimensional evaluation system has been constructed, breaking through the limitation of existing standards that only focus on steady-state reactive power injection. It incorporates response speed, support strength, control accuracy, dynamic quality, and stability performance into a unified evaluation framework, comprehensively quantifying the dynamic reactive power support capability of grid-type inverters and solving the problem of a single evaluation dimension.

[0061] Secondly, a standardized quantitative evaluation model was established. Through an improved VSG dual-mode control strategy and a piecewise linear scoring function, the entire process from simulation testing and index calculation to grade evaluation was standardized. The weights and thresholds of each index were determined based on actual engineering conditions, ensuring the comparability of evaluation results for different equipment and filling the gap in the industry's lack of a unified evaluation method.

[0062] Third, it closely reflects real operating conditions. The evaluation model includes actual factors such as measurement noise, control delay, and harmonic interference. The test signals strictly follow the standards and can accurately reflect the real performance of the equipment under complex power grid disturbances, avoiding the drawbacks of traditional offline testing being disconnected from on-site performance.

[0063] Fourth, it has clear engineering guidance value. The AE five-level scoring system intuitively represents the equipment support capability level, and the control mode of the grid-type photovoltaic inverter is dynamically selected based on the level evaluation results, forming a closed-loop optimization of evaluation-decision-control.

[0064] Example 2: like Figure 6 As shown, Embodiment 2 provides an inverter control system based on multi-dimensional evaluation, including an acquisition module 10, a target determination module 20, a data extraction module 30, an index calculation module 40, an evaluation module 50, and a control module 60. The acquisition module 10 is used to acquire the operating data sequence of the target inverter during a preset time period under fault scenarios; The target determination module 20 is used to determine the reactive power target value of the target inverter based on the preset inverter control model and the system operating parameters of the power grid where the target inverter is located; The data extraction module 30 is used to extract the instantaneous reactive power data sequence of the target inverter during a preset time period, the steady-state reactive power reference value before the fault, the steady-state reactive power average value during the fault, and the reactive power peak value during the fault from the running data sequence. The index calculation module 40 is used to calculate several performance indicators of the target inverter under fault scenarios based on the instantaneous reactive power data sequence, the reactive power target value, the steady-state reactive power reference value, the steady-state reactive power average value, and the reactive power peak value. The evaluation module 50 is used to determine the performance evaluation level of the target inverter based on each of the performance indicators. The control module 60 is used to adjust the current control parameters of the target inverter according to the performance evaluation level, generate optimized control parameters for the target inverter, and control the target inverter according to the optimized control parameters and the inverter control model.

[0065] Furthermore, the acquisition module 10 acquires the operating data sequence of the target inverter during a preset time period under fault scenarios, including: Acquire the initial three-phase voltage data sequence, initial three-phase current data sequence, and initial grid voltage data sequence of the target inverter under a preset time period in a fault scenario, and construct the initial data sequence. Each data sequence in the initial data sequence is downsampled to obtain a running data sequence with a preset time resolution.

[0066] In one possible implementation, the target determination module 20 determines the reactive power target value of the target inverter based on a preset inverter control model and the system operating parameters of the power grid where the target inverter is located, including: Calculate the quotient of the system rated capacity and the system rated voltage in the system operating parameters to obtain the first intermediate value; The grid voltage value at the time of the fault occurrence is obtained from the operating data sequence, and the difference between the system rated voltage and the grid voltage value is calculated to obtain a second intermediate value; The forced injection coefficient of the target inverter under the fault scenario is determined based on the inverter control model. Multiply the first intermediate value, the second intermediate value, and the forced injection coefficient to obtain the reactive power target value of the target inverter.

[0067] In one possible implementation, the data extraction module 30 extracts from the operating data sequence the instantaneous reactive power data sequence of the target inverter over a preset time period, the steady-state reactive power reference value before the fault, the steady-state reactive power average value during the fault, and the reactive power peak value during the fault, including: The three-phase voltage data sequence and the three-phase current data sequence in the running data sequence are respectively subjected to coordinate transformation to obtain the voltage direct axis component data sequence, the voltage quadrature axis component data sequence, the current direct axis component data sequence, and the current quadrature axis component data sequence; Based on the voltage direct-axis component data sequence, the voltage quadrature-axis component data sequence, the current direct-axis component data sequence, and the current quadrature-axis component data sequence, the corresponding instantaneous reactive power data sequence is calculated based on instantaneous reactive power theory. According to the preset first time window, the first subsequence before the fault is extracted from the instantaneous reactive power data sequence, and the steady-state reactive power reference value is calculated based on the first subsequence. According to the preset second time window, a second subsequence during the fault period is extracted from the instantaneous reactive power data sequence, and the steady-state reactive power average value is calculated based on the second subsequence; According to the preset third time window, a third subsequence during the fault period is extracted from the instantaneous reactive power data sequence, and the maximum instantaneous reactive power in the third subsequence is determined as the reactive power peak value.

[0068] In one possible implementation, the indicator calculation module 40 calculates several performance indicators of the target inverter under fault scenarios based on the instantaneous reactive power data sequence, the target reactive power value, the steady-state reactive power reference value, the steady-state reactive power average value, and the reactive power peak value, including: The reactive power response time of the target inverter after a fault occurs is determined based on the steady-state reactive power reference value, the reactive power data sequence, and the reactive power target value. The reactive power support coefficient and steady-state error of the target inverter are calculated based on the steady-state average reactive power value and the target reactive power value. The overshoot of the target inverter is determined based on the peak reactive power and the target reactive power value. Starting from the fault initiation time, the instantaneous reactive power data sequence is traversed, and the first adjustment time that meets the preset conditions is determined. The adjustment time of the target inverter is determined based on the difference between the adjustment time and the fault initiation time. The reactive power response time, the reactive power support coefficient, the steady-state error, the overshoot, and the adjustment time are used as several performance indicators of the target inverter under fault scenarios.

[0069] Furthermore, the preset condition is that within a preset duration after the adjustment time, the instantaneous values ​​of each reactive power in the reactive power data sequence are always within a preset error band.

[0070] In one possible implementation, the evaluation module 50 determines the performance evaluation level of the target inverter based on each of the performance indicators, including: The performance score corresponding to each performance indicator is calculated based on the piecewise function corresponding to each performance indicator. The performance scores are weighted and summed according to preset weighting coefficients to obtain the comprehensive performance score of the target inverter. The performance evaluation level of the target inverter is determined based on the comprehensive performance score.

[0071] In one possible implementation, the control module 60 is configured to adjust the current control parameters of the target inverter according to the performance evaluation level, and generate optimized control parameters for the target inverter, including: If the performance evaluation level is the first level, then the current control parameters of the target inverter will be used as the optimized control parameters of the target inverter. If the performance evaluation level is the second level, then the droop coefficient in the current control parameter is increased and the response time constant in the current control parameter is shortened, thereby generating the optimized control parameters for the target inverter; If the performance evaluation level is level three, the current margin in the current control parameters is increased, the optimized control parameters of the target inverter are generated, and a collaborative support request is sent to the grid dispatch system where the target inverter is located.

[0072] This application provides an inverter control system based on multi-dimensional evaluation. By acquiring operating data sequences under fault scenarios and dynamically determining the reactive power target value in conjunction with grid system parameters, it achieves multi-dimensional quantitative evaluation and closed-loop optimization control of inverter performance. The advantage of this embodiment lies in upgrading the traditional fixed-parameter control strategy to an adaptive mechanism based on real-time feedback, thereby significantly improving the inverter's operational reliability and response accuracy under complex operating conditions. Specifically, by extracting instantaneous reactive power data sequences and key features such as steady-state reference values, average values, and peak values ​​from the operating data sequences, the dynamic response capability of the inverter is comprehensively quantified. Then, performance levels are determined based on multi-dimensional performance indicators, and control parameters are dynamically adjusted according to the performance levels, realizing a shift from passive response to active optimization. This embodiment can not only effectively cope with uncertainties such as photovoltaic output fluctuations and grid impedance changes, but also optimize the inverter's control parameters in real time based on operating data under fault scenarios such as low-voltage ride-through, improving the inverter's subsequent operational reliability and stability.

[0073] For a more detailed explanation of the working principle and procedures of this embodiment, please refer to the relevant description in Embodiment 1.

[0074] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of this application. It should be understood that the above descriptions are merely specific embodiments of this application and are not intended to limit the scope of protection of this application. In particular, it should be noted that any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application for those skilled in the art.

Claims

1. An inverter control method based on multi-dimensional evaluation, characterized in that, include: Obtain the operating data sequence of the target inverter under a preset time period in a fault scenario; Based on the preset inverter control model and the system operating parameters of the power grid where the target inverter is located, the reactive power target value of the target inverter is determined; Extract the instantaneous reactive power data sequence of the target inverter during a preset time period, the steady-state reactive power reference value before the fault, the steady-state reactive power average value during the fault, and the reactive power peak value during the fault from the operating data sequence. Based on the instantaneous reactive power data sequence, the target reactive power value, the steady-state reactive power reference value, the steady-state reactive power average value, and the reactive power peak value, several performance indicators of the target inverter under fault scenarios are calculated. The performance evaluation level of the target inverter is determined based on each of the aforementioned performance indicators; The current control parameters of the target inverter are adjusted according to the performance evaluation level, optimized control parameters of the target inverter are generated, and the target inverter is controlled according to the optimized control parameters and the inverter control model.

2. The inverter control method based on multi-dimensional evaluation as described in claim 1, characterized in that, The acquisition of the target inverter's operating data sequence over a preset time period under fault scenarios includes: Acquire the initial three-phase voltage data sequence, initial three-phase current data sequence, and initial grid voltage data sequence of the target inverter under a preset time period in a fault scenario, and construct the initial data sequence. Each data sequence in the initial data sequence is downsampled to obtain a running data sequence with a preset time resolution.

3. The inverter control method based on multi-dimensional evaluation as described in claim 1, characterized in that, The step of determining the reactive power target value of the target inverter based on the preset inverter control model and the system operating parameters of the power grid where the target inverter is located includes: Calculate the quotient of the system rated capacity and the system rated voltage in the system operating parameters to obtain the first intermediate value; The grid voltage value at the time of the fault occurrence is obtained from the operating data sequence, and the difference between the system rated voltage and the grid voltage value is calculated to obtain a second intermediate value; The forced injection coefficient of the target inverter under the fault scenario is determined based on the inverter control model. Multiply the first intermediate value, the second intermediate value, and the forced injection coefficient to obtain the reactive power target value of the target inverter.

4. The inverter control method based on multi-dimensional evaluation as described in claim 1, characterized in that, The step of extracting the instantaneous reactive power data sequence of the target inverter over a preset time period, the steady-state reactive power baseline value before the fault, the steady-state reactive power average value during the fault, and the reactive power peak value during the fault from the operating data sequence includes: The three-phase voltage data sequence and the three-phase current data sequence in the running data sequence are respectively subjected to coordinate transformation to obtain the voltage direct axis component data sequence, the voltage quadrature axis component data sequence, the current direct axis component data sequence, and the current quadrature axis component data sequence; Based on the voltage direct-axis component data sequence, the voltage quadrature-axis component data sequence, the current direct-axis component data sequence, and the current quadrature-axis component data sequence, the corresponding instantaneous reactive power data sequence is calculated based on instantaneous reactive power theory. According to the preset first time window, the first subsequence before the fault is extracted from the instantaneous reactive power data sequence, and the steady-state reactive power reference value is calculated based on the first subsequence. According to the preset second time window, a second subsequence during the fault period is extracted from the instantaneous reactive power data sequence, and the steady-state reactive power average value is calculated based on the second subsequence; According to the preset third time window, a third subsequence during the fault period is extracted from the instantaneous reactive power data sequence, and the maximum instantaneous reactive power in the third subsequence is determined as the reactive power peak value.

5. The inverter control method based on multi-dimensional evaluation as described in claim 1, characterized in that, The method calculates several performance indicators of the target inverter under fault scenarios based on the instantaneous reactive power data sequence, the target reactive power value, the steady-state reactive power reference value, the steady-state reactive power average value, and the reactive power peak value, including: The reactive power response time of the target inverter after a fault occurs is determined based on the steady-state reactive power reference value, the reactive power data sequence, and the reactive power target value. The reactive power support coefficient and steady-state error of the target inverter are calculated based on the steady-state average reactive power value and the target reactive power value. The overshoot of the target inverter is determined based on the peak reactive power and the target reactive power value. Starting from the fault initiation time, the instantaneous reactive power data sequence is traversed, and the first adjustment time that meets the preset conditions is determined. The adjustment time of the target inverter is determined based on the difference between the adjustment time and the fault initiation time. The reactive power response time, the reactive power support coefficient, the steady-state error, the overshoot, and the adjustment time are used as several performance indicators of the target inverter under fault scenarios.

6. The inverter control method based on multi-dimensional evaluation as described in claim 5, characterized in that, The preset condition is that, within a preset duration after the adjustment time, the instantaneous values ​​of each reactive power in the reactive power data sequence are always within a preset error band.

7. The inverter control method based on multi-dimensional evaluation as described in claim 1, characterized in that, Determining the performance evaluation level of the target inverter based on each of the aforementioned performance indicators includes: The performance score corresponding to each performance indicator is calculated based on the piecewise function corresponding to each performance indicator. The performance scores are weighted and summed according to preset weighting coefficients to obtain the comprehensive performance score of the target inverter. The performance evaluation level of the target inverter is determined based on the comprehensive performance score.

8. The inverter control method based on multi-dimensional evaluation as described in claim 1, characterized in that, The step of adjusting the current control parameters of the target inverter according to the performance evaluation level to generate optimized control parameters for the target inverter includes: If the performance evaluation level is the first level, then the current control parameters of the target inverter will be used as the optimized control parameters of the target inverter. If the performance evaluation level is the second level, then the droop coefficient in the current control parameter is increased and the response time constant in the current control parameter is shortened, thereby generating the optimized control parameters for the target inverter; If the performance evaluation level is level three, the current margin in the current control parameters is increased, the optimized control parameters of the target inverter are generated, and a collaborative support request is sent to the grid dispatch system where the target inverter is located.

9. An inverter control system based on multi-dimensional evaluation, characterized in that, It includes an acquisition module, a target determination module, a data extraction module, an indicator calculation module, an evaluation module, and a control module; The acquisition module is used to acquire the operating data sequence of the target inverter during a preset time period under fault scenarios; The target determination module is used to determine the reactive power target value of the target inverter based on the preset inverter control model and the system operating parameters of the power grid where the target inverter is located; The data extraction module is used to extract from the operating data sequence the instantaneous reactive power data sequence of the target inverter during a preset time period, the steady-state reactive power reference value before the fault, the steady-state reactive power average value during the fault, and the reactive power peak value during the fault. The index calculation module is used to calculate several performance indicators of the target inverter under fault scenarios based on the instantaneous reactive power data sequence, the target reactive power value, the steady-state reactive power reference value, the steady-state reactive power average value, and the reactive power peak value. The evaluation module is used to determine the performance evaluation level of the target inverter based on each of the performance indicators; The control module is used to adjust the current control parameters of the target inverter according to the performance evaluation level, generate optimized control parameters for the target inverter, and control the target inverter according to the optimized control parameters and the inverter control model.

10. The inverter control system based on multi-dimensional evaluation as described in claim 9, characterized in that, The performance index calculation module calculates several performance indicators of the target inverter under fault scenarios based on the instantaneous reactive power data sequence, the target reactive power value, the steady-state reactive power baseline value, the steady-state reactive power average value, and the reactive power peak value, including: The reactive power response time of the target inverter after a fault occurs is determined based on the steady-state reactive power reference value, the reactive power data sequence, and the reactive power target value. The reactive power support coefficient and steady-state error of the target inverter are calculated based on the steady-state average reactive power value and the target reactive power value. The overshoot of the target inverter is determined based on the peak reactive power and the target reactive power value. Starting from the fault initiation time, the instantaneous reactive power data sequence is traversed, and the first adjustment time that meets the preset conditions is determined. The adjustment time of the target inverter is determined based on the difference between the adjustment time and the fault initiation time. The reactive power response time, the reactive power support coefficient, the steady-state error, the overshoot, and the adjustment time are used as several performance indicators of the target inverter under fault scenarios.

Images