A method, medium, and system for evaluating the support capacity of various types of reactive power equipment in a new energy power plant.

Abstract

This invention provides a method, medium, and system for evaluating the reactive power support capability of various types of reactive power regulation equipment in new energy power plants. Belonging to the field of power grid technology, this invention collects operational data from the grid connection point and various reactive power regulation equipment of the new energy power plant, calculates normalized values ​​of technical and economic indicators, maps the voltage time series to a two-dimensional phase space trajectory and calculates the curvature integral of the support trajectory, corrects the normalized values ​​of technical indicators with a geometric support contribution metric, calculates capacity weight, location weight, participation weight, and reliability weight, and then obtains the voltage support contribution index of each piece of equipment. Finally, the contribution indices of all equipment are summed and normalized to output the overall transient voltage support capability evaluation value of the power plant. This invention solves the technical problem that the transient voltage support capability of various types of reactive power regulation equipment cannot be accurately quantitatively evaluated under a unified framework due to heterogeneous physical dimensions, differences in equipment control characteristics, and numerical ill-conditioning of weak power grids.

Description

Technical Field

[0001] This invention belongs to the field of power grid technology, and specifically relates to an evaluation method, medium, and system for the support capability of multiple types of reactive power equipment in new energy power plants. Background Technology

[0002] With the large-scale grid connection of new energy power plants, the coordinated access of multiple types of reactive power regulation equipment, such as static var generators, energy storage converters, and grid-connected converters, has become the main means to ensure the transient voltage stability of the grid connection point. Existing evaluation systems are usually based on single time-domain indicators or expert experience scoring. By extracting scalar features such as reactive power response time and voltage recovery time, and combining them with the analytic hierarchy process (AHP), the voltage support capabilities of various types of equipment are ranked and evaluated. This approach has certain applicability in scenarios involving a single type of equipment.

[0003] However, the above methods have significant limitations when dealing with scenarios where multiple types of equipment coexist. The control characteristics of different equipment cause the voltage response trajectories to exhibit fundamental differences in the time domain, and direct comparison based on scalar indicators will introduce dimensional mismatch and characteristic distortion. At the same time, the high ill-conditioned nature of the node admittance matrix under weak grid conditions will severely amplify the numerical solution error of the location weight, resulting in distortion of the calculation of the voltage contribution of each equipment to the grid connection point.

[0004] Current evaluation methods for renewable energy power plants suffer from several drawbacks. The lack of a unified mathematical framework to eliminate the heterogeneity of equipment control characteristics, coupled with a lack of effective preprocessing techniques for ill-conditioned numerical problems in weak power grids, leads to incomparability in the evaluation results of transient voltage support capabilities for various types of reactive power regulation equipment across different equipment types, resulting in a significant decrease in evaluation accuracy. In other words, existing technologies suffer from the technical problem that the transient voltage support capabilities of various types of reactive power regulation equipment cannot be accurately and quantitatively evaluated within a unified framework due to heterogeneity in physical dimensions, differences in equipment control characteristics, and ill-conditioned numerical problems in weak power grids. Summary of the Invention

[0005] In view of this, the present invention provides a method, medium and system for evaluating the support capability of various types of reactive power equipment in new energy power plants, which can solve the technical problem in the prior art that the transient voltage support capability of various types of reactive power regulation equipment cannot be accurately quantitatively evaluated under a unified framework due to the heterogeneity of physical dimensions, differences in equipment control characteristics and numerical ill-conditioning of weak power grids.

[0006] The present invention is implemented as follows: The first aspect of the present invention provides a method for evaluating the support capability of various types of reactive power equipment in a new energy power plant, comprising the following steps: The system collects operational data from the grid connection points of new energy power plants and various reactive power regulation devices. The operational data includes voltage time series, reactive power time series, reactive current time series, equipment rated capacity, equipment investment cost, and equipment operating time records. Based on the collected operational data, the original values ​​of the indicators for each reactive power regulation device are calculated according to the sets of technical indicators and economic indicators, respectively. The original values ​​of the indicators are then classified into benefit-type indicators or cost-type indicators and subjected to dimensionless normalization processing to obtain normalized indicator values. The collected voltage time series is mapped to a two-dimensional phase space trajectory. The support trajectory curvature integral of the two-dimensional phase space trajectory is calculated. The difference between the support trajectory curvature integrals with and without reactive power regulation equipment is used as the geometric support contribution measure. The normalized index value is corrected based on the geometric support contribution measure, and the normalized index value in the technical index set is updated. A judgment matrix is ​​constructed using an expert method combined with the analytic hierarchy process (AHP). Based on the normalized index values, the weight vectors of technical and economic indicators are calculated. The judgment matrix is ​​then subjected to a consistency test. If the consistency ratio is not less than 0.1, the judgment matrix is ​​reconstructed. After passing the consistency test, the comprehensive score of the transient voltage support capability of a single reactive power regulation device is calculated based on the weight vectors of technical and economic indicators and the updated normalized index values. To address the ill-conditioned problem of node admittance matrix in weak power grid scenarios, the Tikhonov regularization method combined with truncated singular value decomposition is used to preprocess the node admittance matrix. Based on the preprocessing results and collected operating data, the capacity weight, location weight, participation weight, and reliability weight of each reactive power regulation device are calculated. The contribution index of each reactive power regulation device to the voltage support of the substation is calculated by combining the comprehensive score of the transient voltage support capability of a single reactive power regulation device. The equivalent support contribution of the station is obtained by summing the voltage support contribution indices of all reactive power regulation equipment in the station, and the equivalent support contribution of the station is normalized to output the overall transient voltage support capability evaluation value of the station.

[0007] The technical indicator set includes six indicators: reactive response time, reactive voltage regulation coefficient, reactive jump coefficient, voltage recovery time, steady-state voltage deviation, and damping control effect. The economic indicator set includes two indicators: equipment investment cost and voltage improvement cost.

[0008] In the dimensionless normalization process, the normalization formula for the benefit-type indicators is as follows: The normalization formula for cost indicators is: When the sample size of reactive power regulation equipment to be evaluated is insufficient, the 5th percentile and 95th percentile of the historical operating data of no less than 20 reactive power regulation equipment of the same type shall be used as the basis for evaluation. and .

[0009] The two-dimensional phase space trajectory is the voltage time series. The time derivative of the voltage time series Ordered pairs The integral formula for the curvature of the support trajectory of a continuous curve formed on a two-dimensional plane is: The formula for measuring the contribution of geometric support is: .

[0010] Among them, the geometric support contribution metric The normalized index value corresponding to the damping control effect in the set of technical indicators is used to correct the correction coefficient. The correction coefficient is determined by linear regression analysis of no less than 30 sets of simulation and measured comparison experiments. The regression coefficient is taken when the correlation coefficient between the comprehensive score of the transient voltage support capability of the single reactive power regulation equipment after correction and the measured voltage recovery quality is maximized.

[0011] In the construction of the judgment matrix using the expert method combined with the analytic hierarchy process, at least five engineers with experience in the operation and maintenance of new energy power plants conduct pairwise importance comparisons of each indicator, assigning values ​​using a scale of 1 to 9, and calculating the weight of each indicator using the geometric mean method. The consistency test formula is as follows: , ,when The matrix is ​​judged to pass the consistency check.

[0012] The comprehensive scoring formula for the transient voltage support capability of a single reactive power regulation device is as follows: When the primary goal is equipment selection When the value is 0.7, and the primary objective is investment optimization. Take 0.4.

[0013] The regularization solution formula for the Tikhonov regularization method is as follows: Regularization parameters Using the L-curve method to The range is determined by selecting 50 candidate values, and the value corresponding to the inflection point of the L curve is selected. value.

[0014] The truncated singular value decomposition decomposes the nodal admittance matrix into... Keep singular values ​​greater than Singular value components, truncation threshold The stability was determined through numerical stability tests on no fewer than 15 sets of weak grid simulation scenarios.

[0015] Among them, the position weight is obtained through The equivalent impedance ratio is calculated, and the location weight is calculated using the following formula: When the parameters within the power station are incomplete, reactive power voltage sensitivity is used as a substitute, and the normalization formula is as follows: .

[0016] The capacity weight formula is as follows: The participation weighting formula is: The reliability weight formula is: .

[0017] The formula for the voltage support contribution index is as follows: ,satisfy The adjustable weight index is set by engineering requirements or calculated by combining expert methods with the analytic hierarchy process.

[0018] The formula for the equivalent support contribution of the station is as follows: The formula for evaluating the overall transient voltage support capacity of the power station is as follows: , and Composed of no fewer than two stations or no fewer than two operational scenarios in the same batch. The calculation result is obtained by taking the extreme value.

[0019] The formula for the cost of voltage improvement is as follows: The formula for the voltage deficit area is: Voltage improvement .

[0020] The damping control effect is achieved by fitting an exponential decay from the post-fault voltage oscillation peak sequence. Obtain the damping coefficient The formula for damping improvement is: ,in and These are the damping coefficients when the reactive power regulation equipment is connected and not connected, respectively.

[0021] A second aspect of the present invention provides a computer-readable storage medium storing program instructions, which, when executed in a computer, are used to perform the above-described method for evaluating the support capability of multiple types of reactive power equipment in a new energy power plant.

[0022] A third aspect of the present invention provides an evaluation system for the support capability of various types of reactive power equipment in a new energy power plant, comprising the aforementioned computer-readable storage medium, wherein the system is a computer, the computer-readable storage medium is disposed within the system, and the system is provided with a microprocessor for executing program instructions stored in the computer-readable storage medium.

[0023] This invention elevates the voltage dynamic process from a time-domain scalar sequence to a phase-space geometric object. It utilizes the curvature integral of the support trajectory within a Riemannian geometric framework to uniformly measure the two-dimensional phase-space trajectories of various types of reactive power regulation equipment, eliminating evaluation distortion caused by differences in equipment control characteristics. By mapping the voltage trajectories of each device to a unified manifold space, the geometric curvature characteristics are unaffected by differences in specific control strategies. This makes the support capabilities of heterogeneous devices such as static var generators, energy storage converters, and grid-connected converters comparable within the same mathematical framework, thus overcoming the problems of dimensional heterogeneity and characteristic distortion. In summary, this invention solves the technical problem mentioned in the background art where the transient voltage support capabilities of various types of reactive power regulation equipment cannot be accurately quantitatively evaluated within a unified framework due to dimensional heterogeneity, differences in equipment control characteristics, and ill-conditioned numerical conditions in weak power grids. Attached Figure Description

[0024] Figure 1 This is a flowchart of the method of the present invention.

[0025] Figure 2 A comparison chart of the four weights and comprehensive scores of various reactive power regulation devices.

[0026] Figure 3 A comparison diagram of the curvature of the two-dimensional phase space trajectory of each reactive power regulation device. Detailed Implementation

[0027] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings.

[0028] like Figure 1 The diagram shown is a flowchart of an evaluation method for the support capability of multiple types of reactive power equipment in a new energy power plant, provided by the first aspect of this invention. This method includes the following steps: S01. Collect the operating data of the grid connection point of the new energy power station and each reactive power regulation device. The operating data includes voltage time series, reactive power time series, reactive current time series, equipment rated capacity, equipment investment cost and equipment operating time record. S02. Based on the operating data collected in S01, calculate the original values ​​of the indicators for each reactive power regulation device according to the set of technical indicators and the set of economic indicators respectively, and then perform dimensionless normalization processing on the original values ​​of the indicators according to the benefit-type indicators or cost-type indicators to obtain normalized indicator values. S03. Map the voltage time series collected in S01 into a two-dimensional phase space trajectory, calculate the support trajectory curvature integral of the two-dimensional phase space trajectory, use the difference between the support trajectory curvature integrals when the reactive power regulation device is connected and not connected as a geometric support contribution measure, and correct the normalized index value obtained in S02 based on the geometric support contribution measure, and update the normalized index value in the technical index set. S04. Construct a judgment matrix using the expert method combined with the analytic hierarchy process. Calculate the weight vectors of technical and economic indicators based on the normalized index values ​​obtained in S02. Perform a consistency check on the judgment matrix. If the consistency ratio is not less than 0.1, return to reconstruct the judgment matrix. After passing the consistency check, calculate the comprehensive score of the transient voltage support capability of a single reactive power regulation device based on the weight vectors of technical and economic indicators and the normalized index values ​​updated in S03. S05. To address the ill-conditioned problem of the node admittance matrix in weak power grid scenarios, the Tikhonov regularization method combined with truncated singular value decomposition is used to preprocess the node admittance matrix. Based on the preprocessing results and the operating data collected in S01, the capacity weight, location weight, participation weight, and reliability weight of each reactive power regulation device are calculated. Combined with the comprehensive score of the transient voltage support capability of a single reactive power regulation device obtained in S04, the voltage support contribution index of each reactive power regulation device to the substation is calculated. S06. Sum the voltage support contribution indices of all reactive power regulation equipment in the station to obtain the equivalent support contribution of the station, and normalize the equivalent support contribution of the station to output the overall transient voltage support capability evaluation value of the station.

[0029] The technical indicator set includes six indicators: reactive power response time, reactive power voltage regulation coefficient, reactive power jump coefficient, voltage recovery time, steady-state voltage deviation, and damping control effect. The economic indicator set includes two indicators: equipment investment cost and voltage improvement cost. Reactive power response time is the time from when the reactive power regulation equipment begins to respond to reactive power changes after a fault occurs until the output reactive power reaches 90% of the rated reactive power. This reactive power response time is calculated from the reactive current time series collected by S01 and the rated capacity of the equipment collected by S01. The formula for the reactive power voltage regulation coefficient is as follows: In the formula, The actual value of the reactive current output by the reactive power regulation equipment during the fault period is obtained by taking the steady-state value during the fault period from the reactive current time series collected by S01. The initial value of the reactive current output by the reactive power regulation equipment before the fault is obtained by taking the steady-state value before the fault from the reactive current time series collected by S01. The rated reactive current of the reactive power regulating equipment is calculated from the rated capacity of the equipment collected by S01. The actual voltage value at the reactive power regulation equipment terminal during the fault period is obtained by taking the steady-state value during the fault period from the voltage time series collected by S01. The initial voltage value at the reactive power regulation equipment terminal before the fault was obtained by taking the steady-state value before the fault from the voltage time series collected by S01. The rated voltage is obtained from the equipment nameplate parameters in the operating data collected by S01. The formula for the reactive power jump coefficient is as follows: In the formula, The change in reactive power output of the reactive power regulating equipment after a grid voltage disturbance is obtained by taking the difference between the reactive power time series collected by S01 before and after the disturbance. The rated capacity of the reactive power regulating equipment is obtained from the rated capacity of the equipment collected by S01. The voltage change at the reactive power regulation equipment terminal during the fault is obtained by taking the difference between the voltage time series collected by S01 before and after the fault. The formula for voltage recovery time is as follows: In the formula, Voltage time series acquired for S01 The steady-state voltage value after the fault is obtained by averaging the steady-state values ​​from the voltage time series collected by S01. The fault occurrence time is determined by the moment of the first voltage drop in the voltage time series collected by S01. The steady-state voltage deviation formula is expressed as follows: In the formula, The steady-state voltage at the reactive power regulation equipment terminal before the fault is obtained by averaging the steady-state values ​​of the voltage time series collected by S01 before the fault. Source same as above. Damping control effect is obtained from the post-fault voltage oscillation peak sequence. Fit exponential decay Obtain the damping coefficient The formula for the improvement in damping is defined as follows: In the formula, The damping coefficient after connection to the reactive power regulation device is obtained by fitting the voltage time series oscillation peak sequence collected by S01 after connection to the reactive power regulation device. The damping coefficient when the reactive power regulation equipment is not connected is obtained by fitting the voltage time series oscillation peak sequence when the reactive power regulation equipment is not connected.

[0030] The formula for equipment investment cost is expressed as follows: The formula for unit capacity cost is expressed as follows: In the formula, For equipment price, For installation and commissioning costs, For maintenance costs, Other costs are listed above, all obtained itemized from the equipment investment cost data collected by S01. The formula for the voltage deficit area is as follows: In the formula, This is a voltage reference value, obtained from the rated voltage specified in the grid connection standard. The duration of the fault is determined by the start and end times of the voltage drop in the voltage time series collected by S01. The voltage time series was acquired for S01. The formula for voltage improvement is as follows: In the formula, This represents the voltage deficit area when the reactive power regulation equipment is not connected. The voltage deficit area after connecting the reactive power regulation equipment is calculated using the voltage deficit area formula described above. The voltage improvement cost formula is expressed as follows: The smaller the value, the higher the investment efficiency of the reactive power regulation equipment and the lower the voltage improvement cost. The original value of the voltage improvement cost indicator in the set of economic indicators.

[0031] In the dimensionless normalization process, the normalization formulas for the benefit-type indicators (reactive power voltage regulation coefficient, reactive power jump coefficient, and damping control effect) are expressed as follows: The normalized formulas for cost-related indicators (reactive power response time, voltage recovery time, steady-state voltage deviation, equipment investment cost, and voltage improvement cost) are expressed as follows: In the formula, For the first The reactive power regulating equipment in the first The raw values ​​of the indicators under each evaluation metric are derived from the calculation results of various technical and economic indicators. and The values ​​are determined by statistical results from the set of reactive power regulating devices to be evaluated. When the sample size of the set is insufficient, upper and lower limits are set based on engineering experience. These upper and lower limits are obtained through statistical analysis of historical operating data from no fewer than 20 similar reactive power regulating devices. The 5th percentile and 95th percentile of the measured values ​​of each indicator are taken as the upper and lower limits, respectively. and .

[0032] The two-dimensional phase space trajectory is the voltage time series collected by S01. The time derivative of the voltage time series Ordered pairs A continuous curve formed on a two-dimensional plane. The curvature integral of the support trajectory is based on the curvature formula: The integral formula for the curvature of the support trajectory is expressed as follows: In the formula, At the moment when the voltage recovers to steady state, the voltage first enters the steady-state phase in the voltage time series acquired by S01. Time determination of the range for First derivative with respect to time for The second derivative with respect to time; all the above derivatives are obtained by numerical differentiation of the voltage time series acquired by S01. The formula for measuring the contribution of geometric support is as follows: In the formula, For those not connected The integral of the curvature of the support trajectory when the reactive power regulating device is connected is calculated from the voltage time series when the reactive power regulating device is not connected. To access the first The integral of the curvature of the support trajectory after the reactive power regulation device is calculated from the voltage time series collected by S01 after connecting to the reactive power regulation device. A larger value indicates a better transient voltage support effect from the reactive power regulation equipment. Geometric support contribution metric. The normalized index value corresponding to the damping control effect in the technical index set is used to correct the error. The correction coefficient is determined by linear regression analysis of no less than 30 sets of simulation and measured comparative experiments. The regression coefficient that maximizes the correlation coefficient between the comprehensive score of the transient voltage support capability of a single reactive power regulating device after correction and the measured voltage recovery quality is taken. The transient support capability topology measurement algorithm based on the geometric curvature of the voltage trajectory elevates the voltage dynamic process from a time-domain scalar sequence to a phase-space geometric object. It uses the curvature integral of the support trajectory to capture the curvature of the voltage trajectory, transforming the strength of the transient support capability of the reactive power regulating device into the smoothness of the geometric trajectory, thereby avoiding the difficulty of directly comparing indices with different physical dimensions. Furthermore, it uses the Riemannian geometric framework to measure the two-dimensional phase-space trajectory distance of multiple reactive power regulating devices of different types in a unified manifold space, making the support capabilities of different types of reactive power regulating devices such as SVG, energy storage converters, and grid-type converters comparable under the same mathematical framework, eliminating the evaluation distortion problem caused by differences in equipment control characteristics.

[0033] In constructing the judgment matrix using a combination of expert method and analytic hierarchy process, at least five engineers with experience in the operation and maintenance of new energy power plants conducted pairwise importance comparisons of each indicator, assigning values ​​on a scale of 1 to 9 to construct the judgment matrix. ,satisfy The weights of each indicator are calculated using the geometric mean method. The normalized weight formula is expressed as follows: The consistency test formula is expressed as follows: , In the formula, To determine the largest eigenvalue of a matrix, Let be the order of the matrix. The values ​​are for the same order of random consistency, obtained from a table lookup: 0.58 for order 3, 0.90 for order 4, 1.12 for order 5, 1.24 for order 6, 1.32 for order 7, and 1.41 for order 8; when The judgment matrix passes the consistency test, and the resulting weight vector is valid. Technical indicator weight vector. With economic indicator weight vector The scores are calculated using the judgment matrices corresponding to the sets of technical and economic indicators, respectively. The scoring formula for the technical indicators is as follows: The formula for scoring economic indicators is as follows: The formula for comprehensively evaluating the transient voltage support capability of a single reactive power regulation device is as follows: In the formula, The weighting coefficients are based on technical and economic factors, with equipment selection as the primary objective. When the value is 0.7, and the primary objective is investment optimization. The value is set to 0.4. The above typical values ​​are determined by statistical analysis of no less than 10 actual new energy power station engineering cases.

[0034] The Tikhonov regularization method is used to address the ill-conditioned problem of the node admittance matrix in weak power grid scenarios. When the short-circuit ratio of the power grid is less than 2, the condition number of the node admittance matrix is ​​greater than... Direct inversion leads to a significant amplification of numerical errors. The Tikhonov regularization method is used to refine the nodal admittance matrix. After preprocessing, the regularization solution formula is expressed as follows: In the formula, For regularization parameters, The identity matrix; regularization parameter Determined by the L-curve method, for exist to Fifty candidate values ​​with a logarithmic uniform distribution are selected from the range, and the residual norm and solution norm are calculated for each. The value corresponding to the inflection point of the L-curve is then selected. The value, and the inflection point location, are determined by the maximum curvature. Based on regularization and combined with truncated singular value decomposition, the nodal admittance matrix is ​​decomposed into... Keep singular values ​​greater than Singular value components, truncation threshold The maximum truncation ratio, which ensures the location weight calculation error does not exceed 5%, was determined through numerical stability tests on no fewer than 15 sets of weak grid simulation scenarios. The Tikhonov regularization method combined with truncated singular value decomposition (SVD) preprocessing suppresses error amplification in the singular directions of the nodal admittance matrix by introducing a regularization term into the objective function, while simultaneously truncating singular value components with weak numerical significance. This restores the numerical solution of the impedance sensitivity matrix under weak grid conditions to a valid and reliable calculation result, guaranteeing the calculation accuracy of location weights in high-impedance weak grid scenarios.

[0035] The capacity weight formula is expressed as follows: In the formula, For the first The rated capacity of the reactive power regulating equipment is obtained from the rated capacity of the equipment collected by S01. The location weight is obtained by solving the regularization formula. The formula for calculating the equivalent impedance ratio and the location weight is expressed as follows: The normalization formula is expressed as follows: In the formula, The equivalent impedance at the grid connection point, For the first The impedance of the reactive power regulating equipment from the grid connection point is both determined by... The parameters are extracted; when the parameters within the power station are incomplete, reactive power voltage sensitivity is used as a substitute. The formula for reactive power voltage sensitivity is as follows: The normalization formula is expressed as follows: In the formula, To the first The same reactive power change was injected into the reactive power regulating equipment. The change in voltage at the grid connection point is measured during the injection test using the voltage time series acquired by S01. The participation weighting formula is expressed as follows: In the formula, To assess the total number of operating conditions, For the first Under the first working condition The reactive power support provided by the reactive power regulation equipment is obtained from the statistical results of the reactive power time series collected by S01 under various evaluation conditions. The reliability weighting formula is expressed as follows: In the formula, For the first Normal uptime of the reactive power regulating equipment. For the first The total operating time of the reactive power regulation equipment was obtained from the equipment operating time records collected by S01. The formula for the voltage support contribution index is as follows: In the formula, These are adjustable weight indices for capacity weight, location weight, participation weight, and reliability weight, respectively, satisfying... The adjustable weight index is set by engineering requirements or calculated using an expert method combined with the analytic hierarchy process. The formula for the equivalent support contribution of the site is expressed as follows: The formula for evaluating the overall transient voltage support capacity of the power station is expressed as follows: In the formula, and These are the minimum and maximum values ​​of the equivalent support contribution of the stations in the same batch or in the same operational scenario to be compared, respectively, and are calculated based on at least two stations or at least two operational scenarios in the same batch. The calculation result is obtained by taking the extreme value; the overall transient voltage support capability evaluation value of the station. This is the output of S06.

[0036] The specific implementation of step S01 is as follows: Technicians deploy data acquisition devices at the grid connection points of the new energy power plant and at the access nodes of each reactive power regulation device. Voltage time series, reactive power time series, and reactive current time series are synchronously acquired at a sampling frequency of no less than 1000Hz. The sampling duration covers the steady-state period before the fault, the fault duration, and the post-fault recovery period, ensuring that each stage has at least 0.5s of valid data. The rated capacity of the equipment is read from the nameplate parameters. The equipment investment cost is recorded separately for four items: equipment price, installation and commissioning costs, operation and maintenance costs, and other costs. The equipment operating time record is exported from the power plant monitoring system. After acquisition, the raw data is filtered and outlier removal is performed to ensure the integrity and consistency of the data used in subsequent calculations.

[0037] The specific implementation of step S02 is as follows: Based on the operational data collected in S01, technicians calculate the original values ​​of 6 technical indicators and 2 economic indicators according to the calculation formulas for each indicator. The reactive power response time is determined by the reactive current time series, specifying the moment when the equipment outputs 90% of the rated reactive power, and then subtracted from the moment the fault occurred. The reactive voltage regulation coefficient and reactive power jump coefficient are calculated by substituting the steady-state deviation of the reactive current and terminal voltage during the fault into the corresponding formulas. The voltage recovery time is determined by the difference between the moment when the voltage first enters the 5% error band of the steady-state value after the fault in the voltage time series and the moment of the fault. The steady-state voltage deviation is obtained by taking the absolute value of the ratio of the difference between the mean steady-state voltage before and after the fault to the mean steady-state voltage before the fault. The damping control effect is obtained by performing exponential decay fitting on the voltage oscillation peak sequence after the fault to obtain the damping coefficient, and the difference between the damping coefficients for connected and unconnected equipment is calculated as the damping improvement amount. The equipment investment cost is obtained by summing the four costs to obtain the unit capacity cost, and the voltage improvement cost is obtained by the ratio of the voltage deficit area improvement amount to the investment cost. When normalizing, the positive normalization formula is used for benefit-type indicators, and the negative normalization formula is used for cost-type indicators. When the sample is insufficient, the 5% and 95% quantiles of historical data from no less than 20 similar devices are used as the upper and lower bounds for normalization.

[0038] The specific implementation of step S03 is as follows: the voltage time series acquired in S01 is... Its time first derivative Forming an ordered pair A continuous curve drawn on a two-dimensional plane represents the two-dimensional phase space trajectory. The time derivative is calculated using a numerical differentiation method, and a five-point center difference is applied to the sampled data to suppress high-frequency noise. This is based on the curvature formula. Integrating the product of the curvature and the differential of the arc length over the period from the moment of the fault occurrence to the moment the voltage recovers to a steady state, and then dividing by the length of the time window, yields the integral of the curvature of the support trajectory. Calculate the access and non-access rates separately. When the reactive power regulating equipment is in use and The difference between the two This is the geometric support contribution metric. The larger the value, the smoother the voltage trajectory after connecting the device, and the better the transient support effect. As input for the correction coefficient, the normalized index value of the damping control effect in the technical index set is corrected by linear regression analysis (the regression coefficient is determined by no less than 30 sets of simulation and measured comparison experiments), and the normalized index value in the technical index set is updated. The phase space trajectory measurement of different types of equipment is unified by the Riemannian geometric framework to eliminate the influence of control characteristic heterogeneity on the evaluation results.

[0039] The specific implementation of step S04 is as follows: At least five engineers with experience in the operation and maintenance of new energy power stations will conduct pairwise importance comparisons of the six indicators in the technical indicator set, assigning values ​​using a scale from 1 to 9 to construct a 6th-order judgment matrix; and construct a 2nd-order judgment matrix for the two indicators in the economic indicator set. Each judgment matrix satisfies the antisymmetry condition. The geometric mean method is used to calculate the product of elements in each row. The weight vector is obtained by normalizing the root. For a 6th-order judgment matrix, the random consistency index is retrieved from a table. The value is 1.24, so the consistency index is calculated. With consistency ratio ,when Return to reconstruct the judgment matrix until... Until then. After passing the consistency check, based on the technical indicator weight vector... With economic indicator weight vector Calculate the technical indicator scores separately Economic indicators score Then according to the weighting coefficient The weighted average of the transient voltage support capabilities of a single reactive power regulation device is used to obtain a comprehensive score. When the goal is equipment selection When the value is 0.7, and the goal is to optimize investment... Take 0.4.

[0040] The specific implementation of step S05 is as follows: when the grid short-circuit ratio is less than 2, the condition number of the node admittance matrix can exceed... Directly inverting the value would severely amplify the numerical error. Therefore, the Tikhonov regularization method is used, with regularization parameters... Using the L-curve method to Within the range, select 50 candidate values ​​with a logarithmic uniform distribution, calculate the residual norm and solution norm for each, and take the inflection point of the L-curve (determined by the maximum curvature). The value is substituted into the regularization solution formula. The regularized impedance matrix is ​​obtained. Based on this, singular value decomposition is performed on the nodal admittance matrix. Keep singular values ​​greater than The truncation threshold for the component is determined by numerical stability tests of no fewer than 15 sets of weak grid simulation scenarios. Extracting the equivalent impedance of the grid connection point impedance of each device node The following steps are performed: calculating location weight; calculating capacity weight based on equipment rated capacity; calculating participation weight based on reactive power time series statistics for each equipment under various assessment conditions to determine the proportion of reactive power support; and calculating reliability weight based on equipment operating time records. Finally, based on the voltage support contribution index formula, the comprehensive score of each equipment is weighted and combined with the four weight categories to obtain the voltage support contribution index of each reactive power regulation device to the substation. Adjustable weight index The sum of these conditions is 1.

[0041] The specific implementation of step S06 is as follows: The voltage support contribution index of all reactive power regulation equipment within the power station is calculated. Summing yields the equivalent support contribution of the station. To ensure the comparability of evaluation results across different sites or operational scenarios, [the following is a summary of the previous sentence]. Normalization is performed, and the extreme values ​​used for normalization are... and Take samples from no fewer than two stations or no fewer than two operational scenarios within the same batch. The minimum and maximum values ​​of the calculated results. The final output is the overall transient voltage support capability evaluation value of the power station. , The closer the value is to 1, the stronger the overall transient voltage support capability of the power station, which can be directly used as a quantitative basis for the power station's grid connection performance evaluation report.

[0042] It should be noted that the key technologies of this invention include: a phase space measurement technology based on the geometric curvature of voltage trajectories, which elevates the voltage time series from a time-domain scalar series to a continuous geometric object in a two-dimensional phase space. By supporting the integral of trajectory curvature, the curvature of the voltage dynamic process is captured, transforming the strength of the support capability of reactive power regulation equipment with different control characteristics into the smoothness of the geometric trajectory. Furthermore, using a Riemannian geometric framework, distance measurement across equipment types is completed within a unified manifold space, fundamentally eliminating evaluation distortion caused by dimensional heterogeneity and differences in control characteristics; and a joint preprocessing technology combining Tikhonov regularization and truncated singular value decomposition, specifically designed for weak power grids. The problem of amplified numerical errors caused by the high ill-conditioned nature of the node admittance matrix in the scenario is addressed by introducing a regularization term into the objective function to transform the ill-conditioned inversion into a constrained optimization problem. At the same time, singular components with weak numerical significance are truncated, so that the impedance sensitivity matrix can be restored to a numerically stable and reliable result under the condition of high impedance and weak power grid, thus ensuring the accuracy of the location weight calculation. The above two technologies work together to solve the problem of unified measurement of the support capability of multiple types of equipment, and the latter solves the problem of numerical stability in the extraction of topology parameters of weak power grid. Together, they support the ability of this invention to accurately and quantitatively evaluate the transient voltage support capability of multiple types of reactive power regulation equipment in complex engineering scenarios.

[0043] It should be noted that this invention also solves the following technical problem: In the prior art, when multiple reactive power regulation devices are put into operation simultaneously in a new energy power station, it is difficult to separate the actual contribution of each device to the voltage support of the grid connection point, making it impossible for operation and maintenance personnel to identify devices with weak contributions, and thus impossible to optimize scheduling strategies in a targeted manner. This invention constructs a multi-dimensional contribution index system that includes capacity weight, location weight, participation weight, and reliability weight. It uniformly incorporates the rated capacity, network topology location, reactive power support participation under historical operating conditions, and device reliability level of each device into the contribution measurement framework, realizing the quantitative breakdown of the actual contribution share of each device in the overall station voltage support. This enables operation and maintenance personnel to identify device nodes with insufficient contributions, providing a quantitative basis for differentiated operation and maintenance and scheduling optimization of reactive power regulation devices, and solving the technical problem that the contribution of a single device cannot be quantitatively broken down in the scenario of multi-device collaborative operation.

[0044] A second aspect of the present invention provides a computer-readable storage medium storing program instructions, which, when executed in a computer, are used to perform the above-described method for evaluating the support capability of multiple types of reactive power equipment in a new energy power plant.

[0045] A third aspect of the present invention provides an evaluation system for the support capability of various types of reactive power equipment in a new energy power plant, comprising the aforementioned computer-readable storage medium. The system is any one of a computer, a server, or a microcontroller. The computer-readable storage medium is disposed within the system, and the system is provided with a microprocessor that executes the program instructions stored in the computer-readable storage medium.

[0046] Specifically, the principle of this invention is as follows: The core reason why this invention can solve the above-mentioned technical problems lies in the following two aspects. First, after the voltage time series is reconstructed in phase space, the dynamic response differences of different control types of equipment are encoded as the curvature change of the trajectory. As a differential geometric quantity, the curvature integral is not sensitive to the specific form of the control strategy and only reflects the overall curvature of the trajectory. Therefore, it naturally possesses comparability across equipment types. The Riemannian geometric framework further ensures the consistency of distance measurement for trajectories with different topologies in a unified manifold space, making the geometric support contribution measurement an effective tool for eliminating dimensional heterogeneity and characteristic distortion. Second, for weak grid scenarios, Tikhonov regularization, by introducing a regularization term into the objective function, transforms the problem of inverting the ill-conditioned matrix into a constrained optimization problem, suppressing the error amplification effect in singular directions. Truncation of singular value decomposition further eliminates singular components with weak numerical significance, allowing the impedance sensitivity matrix to recover a numerically stable and reliable calculation result under high impedance weak grid conditions, ensuring the calculation accuracy of position weights. The synergistic effect of these two aspects enables this invention to accurately and quantitatively evaluate the transient voltage support capability of multiple types of equipment within a unified framework.

[0047] The following provides a specific embodiment 1 of the present invention, and the specific implementation of each step in this embodiment 1 is described in detail below.

[0048] The specific implementation of step S01 is as follows: Collect operating data from the grid connection point of the new energy power station and each reactive power regulation device. The collected operating data includes voltage time series data. Reactive power time series Reactive current time series Rated capacity of equipment Equipment investment costs The above data, including equipment operating time records, is collected synchronously by the site monitoring system, protection devices, and metering devices. The sampling frequency is not less than 1000Hz, and the collection period covers the steady-state period before the fault, the transient period after the fault, and the recovery period after the fault.

[0049] The specific implementation of step S02 is as follows: Based on the operating data collected in S01, calculate the original values ​​of the technical and economic indicators of each reactive power regulation device. Reactive power response time. The time required for reactive power output to reach 90% of rated reactive power after a fault occurs, determined by... and The calculated reactive power regulation coefficient is expressed as follows: ; In the formula, The steady-state value of reactive current during the fault period is given by... Steady-state phase during fault The initial value of reactive current before the fault is given by... Steady-state phase before failure obtained The rated reactive current is determined by... Converted The steady-state value of the terminal voltage during the fault period is given by... Steady-state segment acquired during fault The initial value of the fault front-end voltage is given by... Steady-state phase before failure obtained The rated voltage is obtained from the parameters on the equipment nameplate. For dimensionless coefficients, their numerators are... This is a dimensionless current ratio, with the denominator being... This is a dimensionless voltage ratio. The formula for the reactive power jump coefficient is as follows: ; In the formula, The change in reactive power before and after the disturbance is denoted by . Obtain the difference before and after the disturbance To affect the voltage change before and after the disturbance, by Obtain the difference before and after the disturbance For dimensionless coefficients, their numerators are... The ratio of dimensionless reactive power, denominator This is a dimensionless voltage ratio. The voltage recovery time formula is as follows: ; In the formula, The average steady-state voltage after the fault is given by... Obtaining the mean value of the steady-state section after the fault The time when the fault occurred, by The timing of the first voltage drop has been determined. and All units are The dimensions on both sides of the equation are unified. The steady-state voltage deviation formula is expressed as follows: ; In the formula, The average steady-state voltage before the fault is given by... Obtaining the mean value of the steady-state period before the fault This is a dimensionless deviation ratio. The damping control effect is assessed by analyzing the peak voltage oscillation sequence after a fault. The damping coefficient is obtained by fitting an exponential decay model. The formula for the fitted model is as follows: ; In the formula, For the first Each voltage oscillation peak, by The oscillation segment after the fault was extracted peak by peak, and the unit is... Consistent is the initial amplitude of the oscillation, and is the fitting parameter, with units of and . Consistent The damping coefficient reflects the rate of oscillation decay and is a dimensionless quantity. The peak index is a positive integer; the above fitting uses the least squares method. Performing linear regression yields Both sides of the equal sign are in the dimension of voltage, thus maintaining a unified dimension. The formula for the improvement in damping is expressed as follows: ; In the formula, The damping coefficient after connecting the reactive power regulation equipment is determined by the following parameters: Oscillating peak sequence fitting obtained The damping coefficient when not connected is obtained by fitting the peak oscillation sequence of the voltage time series when not connected. Since it is a dimensionless quantity, the dimensions on both sides of the equation are unified. The formula for equipment investment cost is expressed as follows: ; In the formula, For equipment price, For installation and commissioning costs, For maintenance costs, Other costs are all obtained from the equipment investment cost data collected by S01, itemized by item. All items are in yuan, and the units on both sides of the equal sign are consistent. The formula for unit capacity cost is expressed as follows: ; In the formula, The unit cost of investment is yuan / unit capacity. The formula for the voltage deficit area is as follows: ; In the formula, The reference value of rated voltage specified in the grid connection standard The duration of the fault is determined by... Determining the start and end times of medium voltage dip Pick and The larger value in, in units of ;integral Units are Divide by (Unit is) )back, It is a dimensionless quantity. The formula for voltage improvement is as follows: ; In the formula, This represents the voltage deficit area when no reactive power regulation equipment is connected. The voltage deficit area after connection is defined; both are dimensionless quantities, and the dimensions on both sides of the equation are unified. The formula for the cost of voltage improvement is as follows: ; In the formula, The unit is yuan / This represents the investment cost required to improve voltage per unit time, and is used only for horizontal comparison and ranking of similar equipment. After normalization, it is used as a dimensionless evaluation index. The smaller the value, the higher the investment efficiency. After completing the initial value calculation, the normalized formulas for the efficiency indicators (reactive power voltage regulation coefficient, reactive power jump coefficient, damping control effect) are expressed as follows: ; The normalized formulas for cost-related indicators (reactive power response time, voltage recovery time, steady-state voltage deviation, equipment investment cost, and voltage improvement cost) are expressed as follows: ; In the formula, For the first Taiwan reactive power regulation equipment The original values ​​of each indicator and The first The maximum and minimum values ​​of each indicator in the set of equipment to be evaluated are determined by statistics of the set of equipment to be evaluated. If the sample is insufficient, the 5% and 95% quantiles of historical operating data from no less than 20 similar devices are used as the values, respectively. and This is the normalized dimensionless index value, ranging from 0 to 1, with the dimensions on both sides of the equal sign being consistent.

[0050] The specific implementation of step S03 is as follows: The voltage time series... Its time derivative Forming an ordered pair This forms a continuous phase space trajectory on a two-dimensional plane. Based on the curvature formula, curvature... The formula is expressed as follows: ; In the formula, for The first numerical derivative with respect to time, in units of for The second numerical derivative with respect to time, in units of The unit is Reflecting the phase space trajectory in The degree of curvature at any given moment, middle The unit is To ensure dimensional uniformity, In this scheme, it is used as a relative metric, and all devices are compared in the same coordinate system. The dimension factor is... The differences cancel each other out in the calculation. The integral formula for the curvature of the support trajectory is expressed as follows: ; In the formula, For the first time the voltage enters The time range, in units of The integration time is expressed in units of 1 / 2. Arc length infinitesimal element middle contain item, It is used only for relative comparisons within the same evaluation system. The formula for measuring the contribution of geometric support is expressed as follows: ; In the formula, For those not connected The integral of the curvature of the support trajectory when the equipment is in use. Curvature integral after connection The larger the value, the better the transient support effect of the equipment; the dimensions on both sides of the equal sign are consistent. The normalized index value used to correct the damping control effect is expressed by the following formula: ; In the formula, To correct the normalized index value of the pre-damping control effect, the normalization process in step S02 is obtained. The regression correction coefficient is determined by linear regression analysis of no less than 30 sets of simulation and measured comparative experiments, and is taken as... The regression coefficient corresponding to the maximum correlation coefficient with the measured voltage recovery quality has an empirical range of 0.1 to 0.3. and These are the sets of equipment to be evaluated in the same batch. Minimum and maximum values This is a dimensionless normalization term, with both sides of the equal sign being dimensionless quantities, thus unifying the dimensions.

[0051] The specific implementation method of step S04 is as follows: no fewer than 5 engineers with experience in the operation and maintenance of new energy power stations compare each indicator pairwise, assign values ​​using a scale of 1 to 9, and construct a judgment matrix. ,satisfy ,in For the first The first indicator is relative to the first The importance ratio of each indicator is a dimensionless quantity. The weight of each indicator is calculated using the geometric mean method, and the formula is expressed as follows: ; In the formula, For the first Each index corresponds to the geometric mean of the elements in each row of the judgment matrix, and is a dimensionless quantity. The normalized weight formula is expressed as follows: ; In the formula, For the first The normalized weights of each indicator are dimensionless, with both sides of the equation having the same dimensions. The consistency test formula is expressed as follows: ; ; In the formula, To determine the largest eigenvalue of a matrix Consistency Indicator Consistency ratio The consistency index for randomness of the same order is obtained from a table lookup: 0.58 for order 3, 0.90 for order 4, 1.12 for order 5, 1.24 for order 6, 1.32 for order 7, and 1.41 for order 8; when If the consistency check is passed, the decision matrix is ​​reconstructed; otherwise, the process is repeated. and All are dimensionless quantities, and the dimensions on both sides of the equals sign are consistent. The weight vector of technical indicators is... The weight vector of economic indicators is The scores are calculated from the judgment matrices corresponding to the sets of technical and economic indicators, respectively. The scoring formula for the technical indicators is as follows: ; In the formula, For the first The normalized weights of each technical indicator are calculated from the technical judgment matrix. The first result obtained in step S02 The first piece of equipment The normalized index values ​​of several technical indicators, among which the damping control effect corresponds to... The correction made in step S03 Alternative Since it is a dimensionless quantity, the dimensions on both sides of the equals sign are unified. The scoring formula for economic indicators is expressed as follows: ; In the formula, For the first The normalized weights of each economic indicator are calculated from the economic judgment matrix; in the economic scoring... Corresponding equipment investment cost indicators Corresponding voltage improvement cost indicators Since it is a dimensionless quantity, the dimensions on both sides of the equation are unified. The comprehensive scoring formula for the transient voltage support capability of a single reactive power regulation device is expressed as follows: ; In the formula, The weighting coefficients are based on technical and economic factors, with equipment selection as the primary objective. The empirical value is 0.7, when investment optimization is the primary objective. The empirical value is 0.4. The above typical values ​​were determined by statistical analysis of no fewer than 10 actual new energy power station engineering cases. For dimensionless comprehensive scoring, the dimensions on both sides of the equal sign are unified.

[0052] The specific implementation of step S05 is as follows: when the grid short-circuit ratio is less than 2, the node admittance matrix... Condition number greater than Direct inversion leads to a significant amplification of numerical errors. Therefore, the Thikhonov regularization method is used to address this issue. After preprocessing, the regularization solution formula is expressed as follows: ; In the formula, Here is the regularization parameter, and here is a dimensionless scalar. identity matrix for transpose The regularized impedance matrix, its dimensions, and the nodal admittance matrix are given. pseudo-inverse consistency Using the L-curve method to The selection is based on 50 candidate values ​​that are logarithmically uniformly distributed within the range, with the value corresponding to the inflection point (maximum curvature) of the L-curve being selected. Value. Combined with truncated singular value decomposition, Decomposed into ,in It is a left singular vector matrix. It is a singular value diagonal matrix. The transpose of the right singular vector matrix retains singular values ​​greater than 1. The amount, for Maximum singular value, truncation threshold The maximum cutoff ratio, determined by numerical stability tests in no fewer than 15 sets of weak grid simulation scenarios, is chosen to ensure that the location weight calculation error does not exceed 5%. The capacity weight formula is expressed as follows: ; In the formula, For the first The rated capacity of the reactive power regulating equipment is obtained from the rated capacity of the equipment collected by S01. For dimensionless weights, the dimensions on both sides of the equation are unified. The formula for calculating positional weights is as follows: ; ; In the formula, The equivalent impedance at the grid connection point, For the first The impedance of the reactive power regulating equipment from the grid connection point is both determined by... Extracted, units are the same For the first The position weighting coefficient of each piece of equipment is a dimensionless quantity. The normalized location weights are dimensionless quantities, with unified dimensions on both sides of the equation. When station parameters are incomplete, reactive power voltage sensitivity is used as a substitute. The formula for reactive power voltage sensitivity is as follows: ; ; In the formula, Indicates the first The reactive voltage sensitivity of the reactive power regulation equipment to the grid connection point is expressed in units of... This reflects the degree of impact of unit reactive power injection on the grid connection point voltage. To the first reactive power change injected into the equipment The change in voltage at the grid connection point is due to Measured during injection testing For the injection of the first The reactive power change of the equipment, in units of All units are the same. Since it is a dimensionless quantity, the dimensions on both sides of the equation are unified. The participation weight formula is expressed as follows: ; In the formula, To assess the total number of operating conditions For the first Under the first working condition The reactive power support provided by the equipment is Statistics obtained from various working conditions dimensionless ratio For dimensionless weights, the dimensions on both sides of the equation are unified. The reliability weight formula is expressed as follows: ; In the formula, For the first Normal uptime of the equipment The total commissioning time is obtained from the operating time records collected by S01, and the units are the same. As a dimensionless weight, the dimensions on both sides of the equation are unified. The formula for the voltage support contribution index is as follows: ; In the formula, , , , Adjustable weighted indices for capacity, location, participation, and reliability, respectively, satisfying... It is obtained from engineering requirements setting or calculation by the analytic hierarchy process (AHP). Since it is a dimensionless quantity, the dimensions on both sides of the equal sign are unified.

[0053] The specific implementation of step S06 is as follows: sum the voltage support contribution indices of all reactive power regulation equipment, and the formula for the equivalent support contribution of the power station is expressed as follows: ; In the formula, Since it is a dimensionless quantity, the dimensions on both sides of the equation are unified. The formula for evaluating the overall transient voltage support capacity of the power station is expressed as follows: ; In the formula, and Each of the following refers to at least two stations or at least two operational scenarios within the same batch. Minimum and maximum values ​​of the calculation results This is a dimensionless evaluation value, ranging from 0 to 1, with the dimensions on both sides of the equals sign being consistent. As the final output.

[0054] To better understand and implement this invention, the following is a specific application scenario example 2: To verify the effectiveness of this invention, technicians set up a test environment and collected operating data from a wind-solar hybrid renewable energy power station's grid connection point and its four reactive power regulation devices. The method steps of this invention were implemented one by one, and evaluation results were output. The power station has a rated installed capacity of 200MW and a grid connection voltage level of 220kV. The four reactive power regulation devices configured in the power station are a static var generator (device A, rated capacity 60Mvar), an energy storage converter (device B, rated capacity 40Mvar), a grid-connected converter (device C, rated capacity 50Mvar), and a dynamic reactive power compensation device (device D, rated capacity 30Mvar). The short-circuit ratio of the power station is 1.8, which is a typical weak grid access scenario.

[0055] In the S01 phase, technicians collected the voltage time series, reactive power time series and reactive current time series of each equipment node during a three-phase short circuit fault at a sampling frequency of 1000Hz. The fault duration was 0.15s, and the acquisition window covered 0.5s before the fault to 2s after the fault. The rated capacity and investment cost data of the equipment were entered from the nameplate parameters and the project ledger.

[0056] In stage S02, the raw values ​​of eight indicators for each device are calculated based on the collected data, as shown in Table 1.

[0057] Table 1 Summary of Original Values ​​of Various Reactive Power Regulation Equipment Indicators

[0058] After normalization, the benefit-type indicators are calculated by forward normalization of the range, and the cost-type indicators are calculated by reverse normalization of the range. The normalized index values ​​of each equipment all fall within the range of 0 to 1.

[0059] In phase S03, the voltage time series of each device node is... Its time derivative Constructing a two-dimensional phase space trajectory, such as Figure 3 As shown, it can be intuitively observed that the phase space trajectory of device C has the smallest curvature and the smoothest trajectory, indicating that its transient support capability is optimal. Figure 3 The broken line in the graph is used to guide the corresponding data distribution. The difference in the integral of the support trajectory curvature between when each device is connected and not connected is calculated to obtain the geometric support contribution measure. Equipment A to D The values ​​were 0.142, 0.198, 0.235, and 0.089, respectively. Based on the linear regression coefficients determined by no fewer than 30 sets of comparative experiments of actual measurements and simulations, the normalized index values ​​of the damping control effect of each device were corrected, and the overall technical performance of device C was further improved after the correction.

[0060] In phase S04, five engineers with experience in the operation and maintenance of new energy power stations compared six technical indicators pairwise, constructed a sixth-order judgment matrix, and calculated the largest eigenvalue. The value is 6.18, which is obtained from the table. It is 1.24. It is 0.036. The value is 0.029, passing the consistency test. In the technical indicator weight vector, the weights are: reactive power response time (0.18), reactive power voltage regulation coefficient (0.22), reactive power jump coefficient (0.19), voltage recovery time (0.17), steady-state voltage deviation (0.14), and damping control effect (0.10). In the economic indicator weight vector, the weights are: equipment investment cost (0.55) and voltage improvement cost (0.45). With equipment selection as the objective... Taking 0.7, the comprehensive score for the transient voltage support capability of a single device is calculated, and the scores for devices A to D are 0.61, 0.74, 0.82, and 0.48, respectively.

[0061] In phase S05, due to the short-circuit ratio of the station being 1.8, the calculated condition number of the node admittance matrix exceeds [a certain threshold]. Direct inversion results in significant errors. The L-curve method is used in... to Filtering regularization parameters within a range Inflection point corresponds to Value Substituting into the Tikhonov regularization formula yields the regularized impedance matrix. Further, perform truncated singular value decomposition on the nodal admittance matrix, retaining singular values ​​greater than 0.5%. The components are calculated, and the effective singular components are retained before extracting the equivalent impedance from each device node to the grid connection point. The calculation results of the four types of weights are shown in Table 2.

[0062] Table 2 Summary Table of Four Categories of Weights for Various Reactive Power Regulation Equipment

[0063] The adjustable weight index is set according to project requirements. Substituting into the voltage support contribution index formula, the voltage support contribution index of equipment A to D is calculated. The values ​​are 0.192, 0.174, 0.241, and 0.072, respectively.

[0064] In phase S06, the equivalent support contribution of the site is obtained by summing the contribution indices of all equipment. It is 0.679, compared with the reference station ( The values ​​are normalized to output the overall transient voltage support capability evaluation value of the power station. The value of 0.655 indicates that the station is at a medium-to-high level among the reference stations in the same batch. Equipment C contributes the most, while equipment D contributes the least. Based on this, maintenance personnel can prioritize optimizing the scheduling strategy for equipment D.

[0065] This invention achieves the following advancements at the principle level compared to traditional methods: Traditional methods rank different types of reactive power regulation equipment based on time-domain scalar indicators, essentially making direct comparisons in a heterogeneous physical dimension space. Because different control strategies for each device lead to variations in the physical meaning of the same scalar indicator, the evaluation results lack inherent consistency. This invention projects the voltage dynamic process onto the phase space geometric manifold and uses curvature integrals—a purely geometric quantity independent of control strategy—to measure support capability, making the evaluation indicators mathematically comparable regardless of equipment type. Furthermore, traditional methods directly calculate position weights by inverting the node admittance matrix in weak grid scenarios, which is severely amplified by the high ill-conditioned nature of the matrix, leading to significant errors in impedance sensitivity calculation. This invention transforms the ill-conditioned inversion problem into a constrained optimization problem through Tikhonov regularization and combines truncated singular value decomposition to eliminate numerically unstable components, making the position weight calculation mathematically well-defined under weak grid conditions, thus ensuring the numerical reliability of the overall evaluation framework in practical engineering scenarios.

[0066] It should be noted that the variables involved in this invention are explained in detail in Tables 3, 4, and 5.

[0067] Table 3. Variable Explanation Table (Part 1)

[0068] Table 4. Variable Explanation Table (Part Two)

[0069] Table 5. Variable Explanation Table (Part 3)

[0070] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for evaluating the support capability of multiple types of reactive power regulation devices in a new energy power station, characterized in that, Includes the following steps: Collect operational data from the grid connection points of new energy power plants and various reactive power regulation equipment; Based on the collected operational data, the original values ​​of the indicators for each reactive power regulation device are calculated according to the sets of technical indicators and economic indicators, and the original values ​​of the indicators are then normalized to obtain normalized indicator values. The collected voltage time series is mapped to a two-dimensional phase space trajectory. The support trajectory curvature integral of the two-dimensional phase space trajectory is calculated. The difference between the support trajectory curvature integrals with and without reactive power regulation equipment is used as the geometric support contribution measure. The normalized index value is corrected based on the geometric support contribution measure, and the normalized index value in the technical index set is updated. A judgment matrix is ​​constructed using the expert method combined with the analytic hierarchy process (AHP), and a consistency check is performed on the judgment matrix. When the consistency ratio is not less than 0.1, the judgment matrix is ​​reconstructed. After passing the consistency check, the comprehensive score of the transient voltage support capability of a single reactive power regulation device is calculated. The node admittance matrix is ​​preprocessed, and the voltage support contribution index of each reactive power regulation device to the station is calculated by combining the comprehensive score of the transient voltage support capability of a single reactive power regulation device. The equivalent support contribution of the station is obtained by summing the voltage support contribution indices of all reactive power regulation equipment in the station, and the equivalent support contribution of the station is normalized to output the overall transient voltage support capability evaluation value of the station.

2. The method of claim 1, wherein the method further comprises: determining the support capability of the new energy power station multi-type reactive power regulation device based on the support capability of the new energy power station multi-type reactive power regulation device. The set of technical indicators includes six indicators: reactive power response time, reactive power voltage regulation coefficient, reactive power jump coefficient, voltage recovery time, steady-state voltage deviation, and damping control effect.

3. The method of claim 2, wherein the method further comprises: determining the support capability of the new energy power station multi-type reactive power regulation device based on the support capability of the new energy power station multi-type reactive power regulation device and the support capability of the new energy power station multi-type reactive power regulation device. The operational data includes voltage time series, reactive power time series, reactive current time series, equipment rated capacity, equipment investment cost, and equipment operating time records.

4. The method of claim 3, wherein the method further comprises: determining the support capability of the new energy power station multi-type reactive power regulation device based on the support capability of the new energy power station multi-type reactive power regulation device and the support capability of the new energy power station multi-type reactive power regulation device. The dimensionless normalization process for the original values ​​of the indicators is specifically performed by classifying them as either benefit-type indicators or cost-type indicators.

5. The method of claim 4, wherein the method further comprises: The geometric support contribution metric is used to correct the normalized index value corresponding to the damping control effect in the technical index set. The correction coefficient is determined by linear regression analysis of no less than 30 sets of simulation and measured comparison experiments. The regression coefficient that maximizes the correlation coefficient between the comprehensive score of the transient voltage support capability of the single reactive power regulation equipment after correction and the measured voltage recovery quality is taken.

6. The method of claim 5, wherein the method further comprises: When constructing the judgment matrix using the expert method combined with the analytic hierarchy process, at least five engineers with experience in the operation and maintenance of new energy power stations conduct pairwise importance comparisons of each indicator, assign values ​​using a scale of 1 to 9, and calculate the weight of each indicator using the geometric mean method.

7. The method of claim 6, wherein the method further comprises: determining the support capability of the new energy power station multi-type reactive power regulation device based on the support capability of the new energy power station multi-type reactive power regulation device. The set of economic indicators includes two indicators: equipment investment cost and voltage improvement cost. 8.The method of claim 7, wherein, The preprocessing of the nodal admittance matrix employs a combination of Tikhonov regularization and truncated singular value decomposition.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores program instructions, which, when executed in a computer, are used to perform the evaluation method for the support capability of multiple types of reactive power equipment in a new energy power plant as described in any one of claims 1-8.

10. An evaluation system for the support capability of multiple types of reactive power equipment in a new energy power plant, characterized in that, The system comprises the computer-readable storage medium of claim 9, wherein the system is a computer, the computer-readable storage medium is disposed within the system, and the system is provided with a microprocessor that executes program instructions stored in the computer-readable storage medium.

Images