A Simulation Method for Small Signal Stability of a New Energy Grid-Connected System

Abstract

The present invention is applicable to the technical field of system stability simulation, and provides a small-signal stability simulation method for a new energy grid-connected system. The method includes: obtaining the dynamic operation parameters of synchronous units and new energy units at each node of the system to generate a set of node dynamic parameters; calculating the inherent equivalent inertia of the synchronous units and the virtual equivalent inertia of the new energy units to generate the total equivalent inertia of the system; constructing an optimization model for virtual equivalent inertia distribution with the new energy penetration rate and the total equivalent inertia of the system as inputs; configuring the optimized virtual equivalent inertia weights into the new energy grid-connected system model, injecting a preset disturbance signal, and performing dynamic simulation to generate dynamic response data; calculating the eigenvalues and damping ratios of the dominant oscillation modes of the system, and quantitatively evaluating the small-signal stability index in combination with the spectral radius of the state matrix; and outputting the optimized virtual equivalent inertia weight configuration scheme and the stability evaluation report to improve the accuracy and reliability of the small-signal stability evaluation of the new energy grid-connected system.

Description

Technical Field

[0001] The present invention relates to the technical field of system stability simulation, and particularly relates to a small-signal stability simulation method for a new energy grid-connected system. Background Art

[0002] Small-signal stability is a key index to measure the ability of a power system to maintain stable operation under small disturbances. With the continuous increase in the proportion of new energy in the power system, the inertia composition and dynamic characteristics of the power system have changed profoundly, and small-signal stability has become a key issue for ensuring the safe operation of the power grid. New energy power generation equipment lacks the mechanical inertia of traditional synchronous generators and is difficult to provide effective inertia support, resulting in an increase in the frequency change rate of the system under power disturbances and a weakening of the oscillation suppression ability, and the risk of small-signal stability has increased significantly.

[0003] In the prior art, the traditional method optimizes the inertia distribution of virtual synchronous generators by establishing a model to improve stability. However, the optimization model does not consider the correction of inertia by multi-source dynamic factors such as load uncertainty, output uncertainty, and fault probability, resulting in insufficient flexibility in inertia calculation. And it only relies on a single index of the minimum damping ratio to judge the small-signal stability of a weak-inertia new energy power system, without considering comprehensive criteria such as the synchronous torque coefficient. These traditional methods are difficult to adapt to the complex and changeable dynamic characteristics in the scenario of high-proportion new energy grid connection, resulting in an incomplete evaluation of small-signal stability and difficulty in meeting different accuracy requirements for simulation parameters. Summary of the Invention

[0004] The present invention provides a small-signal stability simulation method for a new energy grid-connected system, which is used to solve the problems of incomplete evaluation of small-signal stability and difficulty in meeting different accuracy requirements for simulation parameters.

[0005] The present invention provides a small-signal stability simulation method for a new energy grid-connected system, including:

[0006] Obtain the dynamic operation parameters of synchronous units at each node of the system and new energy units at each node of the system, perform node association processing on the dynamic operation parameters, and generate a set of node dynamic parameters including the proportion of the capacity of each unit in the total capacity of the node where it is located;

[0007] Calculate the inherent equivalent inertia of synchronous units and the virtual equivalent inertia of new energy units respectively through the set of node dynamic parameters; wherein, the virtual equivalent inertia is corrected based on the frequency response characteristics of the power grid, and the corrected virtual equivalent inertia and the inherent equivalent inertia are weighted and superimposed according to the proportion of the capacity of each node in the total installed capacity of the system to generate the total equivalent inertia of the system;

[0008] Construct a virtual equivalent inertia distribution optimization model with the new - energy penetration rate and the total equivalent inertia of the system as inputs, and iteratively adjust the virtual equivalent inertia weights and damping coefficients of each node by presetting optimization constraints;

[0009] Configure the optimized virtual equivalent inertia weights into the new - energy grid - connected system model, inject a preset disturbance signal into the new - energy grid - connected system model with the optimized virtual equivalent inertia weights configured, and perform dynamic simulation to generate dynamic response data;

[0010] According to the dynamic response data, calculate the eigenvalues and damping ratios of the system's dominant oscillation mode through eigenvalue analysis, and quantitatively evaluate the small - signal stability index by combining with the spectral radius of the state matrix;

[0011] Output the optimized virtual equivalent inertia weight configuration scheme and the stability evaluation report. The optimized virtual equivalent inertia weight configuration scheme includes the recommended values of virtual equivalent inertia for each node, the optimized damping coefficient, and the corresponding new - energy penetration rate scenarios.

[0012] Furthermore, calculating the inherent equivalent inertia of the synchronous generator set and the virtual equivalent inertia of the new - energy generator set respectively through the node dynamic parameter set includes:

[0013] Inherent equivalent inertia Calculation formula:

[0014]

[0015] Where: is the number of synchronous generator sets, , and are the rated capacity, inertia time constant, and load - fluctuation attenuation coefficient of the rd synchronous generator set respectively, is the fault - probability coefficient of the th synchronous generator set.

[0016] Furthermore, calculating the inherent equivalent inertia of the synchronous generator set and the virtual equivalent inertia of the new - energy generator set respectively through the node dynamic parameter set further includes:

[0017] Virtual equivalent inertia Calculation formula:

[0018]

[0019] Where: is the number of new - energy generator sets, is the virtual inertia control gain of the th new - energy generator set, which is set by the controller parameters, and are the rated capacity and output uncertainty coefficients of the th new energy units respectively, and is the failure probability coefficient of the

[0020] Furthermore, correcting the virtual equivalent inertia based on the grid frequency response characteristics, and weightedly superimposing the corrected virtual equivalent inertia and the inherent equivalent inertia according to the proportion of the capacity of each node in the total installed capacity of the system to generate the total system equivalent inertia, including:

[0021] The corrected virtual equivalent inertia :

[0022]

[0023] Where: is the real-time frequency measurement value, is the grid rated frequency;

[0024] The total system equivalent inertia :

[0025]

[0026] Where: and are the number of nodes of synchronous units and new energy units respectively, and are the inherent equivalent inertia corresponding to the th synchronous unit and the virtual equivalent inertia corresponding to the th new energy unit in each node respectively, and are the rated capacities of the th synchronous unit and the th new energy unit respectively, is the total installed capacity of the system.

[0027] Furthermore, taking the new energy penetration rate and the total system equivalent inertia as inputs, constructing a virtual equivalent inertia allocation optimization model, and iteratively adjusting the virtual equivalent inertia weights and damping coefficients of each node through preset optimization constraints, including:

[0028] The objective function of the virtual equivalent inertia allocation optimization model is to minimize the two-norm of the system state matrix and the weighted frequency response robustness index of the new energy penetration rate , and the expression is as follows:

[0029]

[0030] Wherein: is the system state matrix constructed by the virtual equivalent inertia weight and the damping coefficient ; is the new energy penetration rate, defined as the ratio of the total capacity of new energy units to the total system capacity, and are the weight coefficients, is the frequency response correction factor, is the frequency deviation of the sensitivity matrix to the disturbance frequency ; is the matrix infinity norm.

[0031] Furthermore, taking the new energy penetration rate and the total equivalent inertia of the system as inputs, constructing a virtual equivalent inertia allocation optimization model, and iteratively adjusting the virtual equivalent inertia weights and damping coefficients of each node through preset optimization constraints, further includes:

[0032] The constraint conditions of the virtual equivalent inertia allocation optimization model include virtual equivalent inertia weight constraints of nodes, damping coefficient range constraints, and stability criterion threshold constraints. Among them, the stability criterion thresholds include the real part of the eigenvalue, the damping ratio, and the synchronizing torque coefficient.

[0033] Furthermore, configuring the optimized virtual equivalent inertia weight into the new energy grid-connected system model, injecting a preset disturbance signal into the new energy grid-connected system model with the optimized virtual equivalent inertia weight configured, and performing dynamic simulation to generate dynamic response data, includes:

[0034] Configuring the optimized virtual equivalent inertia weight into the corresponding node controller in the new energy grid-connected system model, and updating the inertia distribution parameters of the system;

[0035] Injecting a preset disturbance signal into the new energy grid-connected system model with the optimized virtual equivalent inertia weight configured, performing time-domain dynamic simulation, and recording the frequency deviation curve of each node and the change curve of the power angle difference between synchronous units in real time to generate frequency response data and power angle oscillation characteristic data;

[0036] From the frequency response data, calculating the difference between the maximum frequency and the steady-state frequency, and determining the frequency overshoot by dividing the difference by the steady-state frequency; and calculating the time from the start of the disturbance to when the frequency returns to within the steady-state error band as the adjustment time; extracting the maximum instantaneous absolute value of the power angle difference of the synchronous units from the power angle oscillation characteristic data as the maximum power angle swing.

[0037] Furthermore, injecting a preset disturbance signal into the new energy grid-connected system model with the optimized virtual equivalent inertia weight configured, includes:

[0038] The preset disturbance signal includes at least one of a step load disturbance or a random power fluctuation disturbance;

[0039] The step load disturbance is realized by instantaneously increasing or decreasing the load at a specified bus, with the change amplitude being a preset percentage of the total system capacity and the duration being set as a short-term transient process;

[0040] The random power fluctuation disturbance includes injecting a random power fluctuation sequence conforming to a statistical distribution at the grid connection point of a new energy unit, and the fluctuation amplitude is determined based on the percentage of the rated capacity of the new energy unit.

[0041] Furthermore, based on the dynamic response data, calculating the eigenvalues and damping ratios of the system's dominant oscillation mode through eigenvalue analysis, and combining with the spectral radius of the state matrix to quantitatively evaluate the small-signal stability index, including:

[0042] Constructing a state-space equation including the rotor dynamic equation of a synchronous unit, the virtual equivalent inertia control logic of a new energy unit, and the grid network equation based on the new energy grid-connected system model with the optimized virtual equivalent inertia weight configured;

[0043] Calculating the eigenvalues of the state matrix in the state-space equation, and screening out the low-frequency oscillation modes within a preset frequency range as the dominant oscillation modes;

[0044] Calculating the damping ratio of the dominant oscillation mode, where the damping ratio reflects the attenuation speed of the oscillation amplitude; at the same time, calculating the spectral radius of the state matrix, and the spectral radius characterizes the maximum increase amplitude of the system's dynamic response;

[0045] If the damping ratio reaches or exceeds the preset lower limit and the spectral radius is less than or equal to the preset radius, it is determined that the small-signal stability of the system meets the standard;

[0046] Otherwise, it is determined as not meeting the standard, and the virtual equivalent inertia weight and damping coefficient are re-optimized until the stability requirement is met.

[0047] Furthermore, outputting the optimized virtual equivalent inertia weight configuration scheme and the stability evaluation report, and the optimized virtual equivalent inertia weight configuration scheme includes the recommended values of the virtual equivalent inertia for each node, the optimized damping coefficient, and the corresponding new energy penetration scenarios, including:

[0048] Based on the optimized virtual equivalent inertia weight and damping coefficient, generating the recommended values of the virtual equivalent inertia for each node, binding the recommended values of the virtual equivalent inertia to the new energy penetration scenarios, and verifying the effectiveness according to the frequency response data and power angle oscillation characteristic data in the dynamic simulation;

[0049] Extract the damping ratio of the dominant oscillation mode, the spectral radius of the state matrix, and the real part distribution of the eigenvalues, and generate the stability level evaluation result in combination with the preset threshold;

[0050] The stability evaluation report also includes the comparative analysis of the virtual equivalent inertia distribution under different permeability scenarios and the stability risk warning suggestions.

[0051] It can be seen from the above technical solutions that the present invention has the following advantages:

[0052] After obtaining the dynamic operation parameters of the synchronous units and new energy units, the present invention processes them to generate a node dynamic parameter combination including the proportion of the capacity of each unit in the total capacity of the corresponding node, ensuring that the subsequent inertia calculation closely conforms to the actual operation characteristics of the node; calculates the inherent equivalent inertia of the synchronous units and the virtual equivalent inertia of the new energy units respectively, and superimposes them to generate the total system equivalent inertia after correcting the virtual equivalent inertia in combination with the grid frequency response characteristics, fully integrating the inertia characteristics of the two types of units, and the total system equivalent inertia can truly reflect the system inertia level; constructs a virtual equivalent inertia distribution optimization model with the new energy penetration rate and the total system equivalent inertia as inputs, iteratively adjusts the virtual equivalent inertia weight and damping coefficient through preset constraint conditions, configures the optimized virtual equivalent inertia weight into the system model and injects a preset disturbance signal for dynamic simulation to generate dynamic response data, providing data support for analyzing the dynamic characteristics of the system under real disturbances by simulating the disturbance scenarios in actual operation; calculates the eigenvalues and damping ratio of the dominant oscillation mode through eigenvalue analysis, and quantifies and evaluates the small-signal stability index in combination with the spectral radius of the state matrix, avoiding the limitations of a single index through a multi-dimensional analysis method; finally outputs a configuration plan and a stability evaluation report including the recommended value of the virtual equivalent inertia, the optimized damping coefficient, and the permeability scenario. The present invention effectively improves the accuracy and reliability of the small-signal stability evaluation of the new energy grid-connected system, and provides comprehensive data support for the safe and stable operation of the power grid under different new energy penetration scenarios. Description of the Drawings

[0053] Figure 1 It is a schematic flowchart of the embodiment of a small-signal stability simulation method for a new energy grid-connected system in the present invention;

[0054] Figure 2 It is a schematic diagram of the simulation system structure in the present invention. Detailed Embodiments

[0055] In the description of the present application and the above-mentioned drawings, terms such as "first", "second", "third", "fourth", etc. (if any) are used to distinguish similar objects and do not necessarily describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances so that the embodiments of the present application described here can be implemented in an order other than those illustrated or described here. In addition, the terms "comprising" and "corresponding to" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device comprising a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0056] Embodiment 1

[0057] In this embodiment, the implementation method can be implemented in a system, in a server, or in a terminal, and no specific limitation is made. Below, from the perspective of system implementation, the small-signal stability simulation method for a new energy grid-connected system in the present application will be introduced. Please refer to Figure 1 , the method provided by the embodiment of the present application includes the following steps:

[0058] S11. Obtain the dynamic operation parameters of the synchronous units and new energy units at each node of the system, perform node association processing on the dynamic operation parameters, and generate a set of node dynamic parameters including the proportion of the capacity of each unit in the total capacity of the node where it is located;

[0059] Taking the Guangxi Power Grid as an example, export the dynamic stability data of the synchronous units from the SCADA (Supervisory Control And Data Acquisition) data acquisition and monitoring control system, including the unit capacity, inertia time constant, and historical failure rate, as well as the grid-connected bus number; obtain the virtual equivalent inertia control parameters, output prediction error coefficients, and equipment availability rates of wind power and photovoltaic units from the new energy energy management platform, and record their grid-connected bus numbers. At the same time, parse the topology structure of the entire network AC system based on the common information model file of the Guangxi Power Grid, extract the bus node list and connection relationship, and generate a node-unit association index table.

[0060] Intelligently match the grid-connected bus numbers of the synchronous units and new energy units with the node numbers in the power grid topology. For example, directly associate buses with unified coding through the number field; bind the low-voltage buses accessed by distributed new energy units to the nearest nodes in the topology model using geographical location coordinates. Secondly, check and complete the data, such as handling missing parameters and eliminating abnormal data.

[0061] Associate the scattered unit dynamic operation parameters with the physical nodes in the power grid topology: For each node, accumulate the capacities of all synchronous units and new energy units under it to obtain the total node capacity; summarize the capacities of all nodes to generate the total installed capacity of the system; calculate the proportion of each unit's capacity to the node capacity and write it into the node dynamic parameter set. The finally generated parameter set is a structured table, with each row corresponding to a node and containing the following fields: node number, unit type (synchronous / new energy), unit capacity, inertia time constant (for synchronous units), virtual inertia control parameter (for new energy units), failure probability coefficient, output uncertainty coefficient (for new energy units), capacity proportion.

[0062] S12. Calculate the inherent equivalent inertia of synchronous units and the virtual equivalent inertia of new energy units respectively through the node dynamic parameter set; among them, correct the virtual equivalent inertia based on the power grid frequency response characteristics, and perform weighted superposition of the corrected virtual equivalent inertia and the inherent equivalent inertia according to the proportion of each node's capacity to the total installed capacity of the system to generate the total equivalent inertia of the system;

[0063] 1. Extract the rated capacity, inertia time constant, load fluctuation attenuation coefficient, and failure probability coefficient of synchronous units from the node dynamic parameter set; among them, the load fluctuation attenuation coefficient is dynamically adjusted based on the historical load deviation rate, specifically determined according to the ratio of the load fluctuation standard deviation of the node where the unit is located in the past year to the regional average fluctuation level, and the value range is . The failure probability coefficient is assigned through the equipment reliability model or historical failure records. For example, the proportion of the annual failure times of a certain unit, the higher the unit failure probability, the worse its inertia availability.

[0064] For each synchronous unit, calculate its inherent equivalent inertia contribution value, summarize the contribution values of all synchronous units, and generate the total inherent equivalent inertia of the system :

[0065]

[0066] Among them: is the number of synchronous units, , and are the rated capacity, inertia time constant, and load fluctuation attenuation coefficient of the rd synchronous unit respectively, is the failure probability coefficient of the th synchronous unit.

[0067] 2. Extract the virtual inertia control gain, rated capacity, output uncertainty coefficient, and fault probability coefficient of new energy units from the dynamic parameter set of slave nodes. Among them, the output uncertainty coefficient is calculated based on the root mean square error between the actual output and the predicted output; the fault probability coefficient is assigned according to the equipment maintenance records or the mean time between failures (MTBF) data provided by the manufacturer. The output uncertainty coefficient is calculated as follows:

[0068]

[0069] Where: is the number of samples within the statistical time window, and are respectively the actual output and the predicted output at time

[0070] Calculate the virtual equivalent inertia contribution value for each new energy unit, sum up the contribution values of all new energy units, and generate the total system virtual equivalent inertia :

[0071]

[0072] Where: is the number of new energy units, is the virtual inertia control gain of the th new energy unit, set by the controller parameters, and are respectively the rated capacity and the output uncertainty coefficient of the th new energy unit, is the th new energy unit's fault probability coefficient

[0073] 3. Obtain the real-time frequency of the power grid through a phasor measurement unit (PMU) , calculate the frequency deviation , where is the rated frequency; dynamically adjust the virtual equivalent inertia according to the frequency deviation:

[0074]

[0075] Where: , set the upper limit of the correction coefficient to 1.2, set the lower limit of the correction coefficient to 0.8

[0076] Weight and superimpose the corrected equivalent inertia according to the proportion of the capacity of each node in the total installed capacity of the system to generate the total system equivalent inertia as follows:

[0077]

[0078] Wherein: and are the number of nodes of the synchronous generator set and the new energy generator set respectively, and are respectively the inherent equivalent inertia corresponding to the th synchronous generator set and the virtual equivalent inertia corresponding to the th new energy generator set in each node, and are respectively the th synchronous generator set and the th new energy generator set rated capacity, is the total installed capacity of the system.

[0079] Finally, the total equivalent inertia of the system is input into the simulation model to verify whether the frequency response curve meets the requirements of the preset overshoot (≤5%) and settling time (≤10 s).

[0080] The above steps dynamically correct the virtual equivalent inertia through the frequency deviation to enhance the inertia support ability of the system in the frequency emergency state; the capacity ratio weighting ensures the dominant role of the large-capacity generator set in the system inertia.

[0081] S13. With the new energy penetration rate and the total equivalent inertia of the system as inputs, construct a virtual equivalent inertia allocation optimization model, and iteratively adjust the virtual equivalent inertia weights and damping coefficients of each node by presetting optimization constraints;

[0082] 1. By optimizing the virtual equivalent inertia weights and damping coefficients, minimize the system dynamic response risk and improve the frequency robustness, as follows:

[0083] The objective function of the virtual equivalent inertia allocation optimization model is to minimize the two-norm of the system state matrix and the new energy penetration rate weighted frequency response robustness index , and the expression is as follows:

[0084]

[0085] Wherein: is the system state matrix constructed by the virtual equivalent inertia weight and the damping coefficient , is the new energy penetration rate, is defined as the ratio of the total capacity of the new energy generator set to the total capacity of the system, and are weight coefficients, , used to balance the priority of the state matrix stability and the frequency robustness, is the frequency deviation is the sensitivity matrix to the disturbance frequency , is the matrix infinity norm, representing the frequency deviation suppression ability under the worst disturbance. The first term in the formula : The larger the total equivalent inertia of the system, the larger the allowable fluctuation of the system dynamic response. It is necessary to reduce the state matrix gain to maintain stability; the second term : The higher the new energy penetration rate, the more critical the frequency anti-disturbance ability, and it is necessary to focus on optimization.

[0086] 2. The constraint conditions of the virtual equivalent inertia distribution optimization model include the virtual equivalent inertia weight constraint of the node, the damping coefficient range constraint, and the stability criterion threshold constraint. Among them, the stability criterion threshold includes the real part of the eigenvalue, the damping ratio, and the synchronizing torque coefficient.

[0087] Specifically, the virtual equivalent inertia weight constraint of the node:

[0088]

[0089]

[0090] Among them: and are the virtual equivalent inertia weight and capacity of the th new energy unit. The upper limit of the single-node weight is that the virtual equivalent inertia distribution of the new energy unit does not exceed its capacity ratio to avoid over-allocation; the lower limit of the total weight is that the total contribution of the virtual equivalent inertia needs to match the new energy penetration rate and the total inertia of the system to ensure the inertial support demand; is the virtual equivalent inertia distribution ratio coefficient, which is set by the system inertia demand.

[0091] Damping coefficient range constraint:

[0092]

[0093] Among them: is the damping coefficient of the th unit, and are the preset lower and upper limits of the damping coefficient respectively. The damping coefficient needs to be within a reasonable range to prevent over-damping (slow response) or under-damping (continuous oscillation).

[0094] Stability criterion threshold constraint:

[0095]

[0096] Among them: is the real part of the eigenvalue of the dominant oscillation mode, which needs to be less than the negative threshold , ensure rapid decay of oscillations; is the damping ratio and needs to be greater than the lower limit , suppress the oscillation amplitude; is the synchronizing torque coefficient and needs to be positive to maintain the rotor angle stability.

[0097] 3. Iterative optimization process:

[0098] Set the initial virtual equivalent inertia weight , damping coefficient ; construct the state matrix , calculate the objective function ; use the gradient descent algorithm or particle swarm optimization (PSO) to calculate the gradients of the objective function with respect to and , and update the parameters:

[0099]

[0100]

[0101] where: is the learning rate.

[0102] Apply constraints to the updated parameters: truncate to , limit within . Verify the stability criterion. If not met, return and continue the iteration. When the change rate of the objective function and the stability criterion are satisfied, terminate the optimization.

[0103] S14. Configure the optimized virtual equivalent inertia weight into the new - energy grid - connected system model, inject a preset disturbance signal into the new - energy grid - connected system model configured with the optimized virtual equivalent inertia weight, and perform dynamic simulation to generate dynamic response data;

[0104] 1. Configure the optimized virtual equivalent inertia weight into the node controller corresponding to the new - energy grid - connected system model, and update the inertia distribution parameters of the system;

[0105] The inertia distribution parameters of the system refer to the virtual equivalent inertia weights and damping coefficients of the virtual synchronous generators (VSGs) at each node; the weight conversion is to convert the optimized weight into the actual inertia time constant; the damping coefficient assignment is to directly use the optimized . After the parameter update, check the VSG module parameters in the Simulink model to confirm that and have been updated. Please refer to Figure 2 , Sync - G node (1 - 3): Keep the fixed parameters , , no need to update; VSG nodes (4, 7, 9): update parameters according to the optimization results.

[0106] 2. Inject a preset disturbance signal into the new energy grid-connected system model configured with the optimized virtual equivalent inertia weight, perform time-domain dynamic simulation, record the frequency deviation curves of each node and the change curves of the power angle difference between synchronous units in real time, and generate frequency response data and power angle oscillation characteristic data;

[0107] Specifically, the preset disturbance signal includes at least one of step load disturbance or random power fluctuation disturbance; the step load disturbance is realized by instantaneously increasing or decreasing the load at a specified bus, the change amplitude is a preset percentage of the total system capacity, and the duration is set as a short transient process; the random power fluctuation disturbance includes injecting a random power fluctuation sequence conforming to a statistical distribution at the new energy unit connection point, and the fluctuation amplitude is determined based on the percentage of the new energy unit rated capacity.

[0108] Such as Figure 2 , the step load disturbance can be set at all load nodes (Load5, Load6, Load8). For example, a 10% sudden drop of the total system capacity (such as a 100 MW sudden drop in a 1000 MW system) is applied at Load5, lasting for 0.1 seconds. The random power fluctuation disturbance can be set at all or some of the VSG nodes (VSG4, VSG7, VSG9). For example, a normal distribution random fluctuation with an amplitude of 5% of the rated capacity (such as a ±10 MW fluctuation for a 200 MW unit) is injected at VSG4. Simulation execution: Set simulation parameters: Time: 0 - 10 seconds; Step size: 1 ms; Solver: ode23t (suitable for stiff systems). Frequency deviation: of Sync-G nodes (1 - 3) ; Power angle difference: .

[0109] 3. From the frequency response data, calculate the difference between the maximum frequency and the steady-state frequency, and determine the frequency overshoot by dividing the difference by the steady-state frequency; and calculate the time from the start of the disturbance to when the frequency returns to within the steady-state error band as the regulation time; extract the maximum instantaneous absolute value of the synchronous unit power angle difference from the power angle oscillation characteristic data as the maximum power angle swing.

[0110] By analyzing the frequency response data, calculate the difference between the maximum frequency and the steady-state frequency, and determine the frequency overshoot by dividing the difference by the steady-state frequency. And calculate the time from the start of the disturbance to when the frequency returns to within the steady-state error band as the regulation time; identify the maximum deviation value from the power angle oscillation data as the power angle swing to quantify the system dynamic response performance. This process combines the peak detection algorithm and the time window determination to ensure the accuracy and repeatability of index extraction.

[0111] S15. Calculate the eigenvalues and damping ratios of the system's dominant oscillation modes through eigenvalue analysis based on the dynamic response data, and quantitatively evaluate the small-signal stability index in combination with the spectral radius of the state matrix.

[0112] 1. Based on the new energy grid-connected system model with the optimized virtual equivalent inertia weight configured, construct a state-space equation that includes the rotor dynamic equation of the synchronous unit, the virtual equivalent inertia control logic of the new energy unit, and the grid network equation.

[0113] Synchronous unit model (nodes 1-3): The dynamic behavior of the synchronous generator is described by the rotor motion equation:

[0114]

[0115] Where: is the rotor angle of the th synchronous generator, is the speed deviation of the th synchronous generator, is the initial speed deviation, is the inertia time constant of the th synchronous generator, is the damping coefficient of the th synchronous generator, representing the attenuation ability of the speed deviation, and are the mechanical power input and electromagnetic power output of the th synchronous generator.

[0116] VSG control model (nodes 4, 7, 9): The dynamic equation of the virtual synchronous generator (VSG) simulates the inertial characteristics of the synchronous unit:

[0117]

[0118] Where: is the inertia time constant of the th VSG, is the damping coefficient of the th VSG, and are the reference power and actual output power of the th VSG respectively.

[0119] Power balance equation based on the nodal admittance matrix:

[0120]

[0121] Where: 、 are the active and reactive powers injected at node , is the voltage phasor of the node , and is the element of the nodal admittance matrix, which characterizes the topological connection relationship of the power grid.

[0122] 2. Calculate the eigenvalues of the state matrix in the state space equation, and screen the low-frequency oscillation modes within the preset frequency range as the dominant oscillation modes;

[0123] Linearize the synchronous generator set and VSG equations to generate a 10×10 state matrix A, which includes: 3 synchronous generator sets (2 state variables for each set , ), and 3 VSGs (2 state variables for each set , ).

[0124] Calculate the eigenvalues of the state matrix A , and screen the low-frequency oscillation modes:

[0125]

[0126] Among them: is the real part of the eigenvalue, which reflects the oscillation decay rate, is the imaginary part of the eigenvalue, corresponding to the oscillation frequency, is the oscillation frequency.

[0127] 3. Calculate the damping ratio of the dominant oscillation mode, and the damping ratio reflects the attenuation speed of the oscillation amplitude; at the same time, calculate the spectral radius of the state matrix, and the spectral radius characterizes the maximum increase of the system dynamic response;

[0128] 4. If the damping ratio reaches or exceeds the preset lower limit and the spectral radius is less than or equal to the preset radius, it is determined that the small-signal stability of the system meets the standard;

[0129] 5. Otherwise, it is determined to be unqualified, and the virtual equivalent inertia weight and damping coefficient are re-optimized until the stability requirements are met.

[0130] First, based on the constructed state matrix A, calculate its eigenvalues and screen out the dominant low-frequency oscillation modes, and calculate the damping ratio through the formula : , and at the same time obtain the spectral radius , where is the eigenvalue of the dominant oscillation mode, and represent the real part and the imaginary part of the eigenvalue respectively; if the damping ratio satisfies and the spectral radius , then it is determined that the system is stable; otherwise, the virtual equivalent inertia weight and damping coefficient need to be adjusted by the gradient descent or particle swarm algorithm, and after updating the parameters, the simulation and stability evaluation are re-executed until all the indicators meet the standard.

[0131] S16. Output the optimized virtual equivalent inertia weight configuration scheme and stability assessment report. The optimized virtual equivalent inertia weight configuration scheme includes the recommended values of virtual equivalent inertia for each node, the optimized damping coefficient, and the corresponding new energy penetration scenarios.

[0132] 1. Based on the optimized virtual equivalent inertia weight and damping coefficient, generate the recommended values of virtual equivalent inertia for each node. The recommended values are bound to the new energy penetration scenarios, and their effectiveness is verified according to the frequency response data and power angle oscillation characteristic data in the dynamic simulation.

[0133] 2. Extract the damping ratio, state matrix spectral radius, and real part distribution of eigenvalues of the dominant oscillation mode, and generate the stability level assessment results in combination with the preset threshold.

[0134] 3. The report also includes the comparative analysis of virtual equivalent inertia distribution under different penetration scenarios and the stability risk warning suggestions.

[0135] Specifically, based on the optimized virtual equivalent inertia weight and damping coefficient, generate the recommended values of virtual equivalent inertia for each node. For example, for node 4 , and for node 7 ; and bind them to different new energy penetration scenarios (such as 30%, 50%, 70%). Verify their effectiveness through the frequency overshoot and power angle swing in the dynamic simulation. Extract the damping ratio, state matrix spectral radius, and real part distribution of eigenvalues of the dominant oscillation mode, and generate the stability level (such as "excellent", "warning", "dangerous"). Finally, output a report to compare the inertia distribution differences under different penetration scenarios (such as higher is required for high penetration), and put forward risk warnings (such as additional standby inertia is required when the penetration exceeds 50%) to ensure that the configuration scheme can directly guide the engineering implementation.

[0136] Through the cooperation of multiple steps in the above embodiments, the implementation realizes data integration, inertia calculation, optimized distribution, simulation evaluation, and result output. Accurately process multi-source dynamic data, comprehensively consider the system inertia characteristics and new energy penetration rate, optimize the virtual equivalent inertia distribution, and comprehensively evaluate the stability, effectively making up for the deficiencies of traditional methods in terms of uncertainty adaptation, multi-scenario adaptation, and comprehensive evaluation. Effectively improve the accuracy and reliability of the small-signal stability assessment of the new energy grid-connected system, and provide technical support for the safe and stable operation of the power grid under different new energy penetration scenarios.

[0137] It can be understood that those skilled in the art can, under the guidance of the above embodiments, combine various implementation manners in the above embodiments to obtain technical solutions of various implementation manners.

[0138] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A small-signal stability simulation method for a new energy grid-connected system, characterized in that, Including: Obtain the dynamic operation parameters of the synchronous units and new energy units at each node of the system, perform node association processing on the dynamic operation parameters, and generate a node dynamic parameter set including the proportion of the capacity of each unit in the total capacity of the corresponding node; Calculate the inherent equivalent inertia of the synchronous units and the virtual equivalent inertia of the new energy units respectively through the node dynamic parameter set; Inherent equivalent inertia Calculation formula: Wherein: is the number of synchronous units, , and are respectively the rated capacity, inertia time constant and load fluctuation attenuation coefficient of the th synchronous unit, is the failure probability coefficient of the th synchronous unit; Virtual equivalent inertia Calculation formula: Wherein: is the number of new energy units, is the virtual inertia control gain of the th new energy unit, which is set by the controller parameters, and are respectively the rated capacity and output uncertainty coefficient of the th new energy unit, and is the failure probability coefficient of the th new energy unit; Among them, the virtual equivalent inertia is corrected based on the grid frequency response characteristics, and the corrected virtual equivalent inertia and the inherent equivalent inertia are weighted and superimposed according to the proportion of the capacity of each node in the total installed capacity of the system to generate the total equivalent inertia of the system; Modified virtual equivalent inertia : Wherein: is the real-time frequency measurement value, is the rated frequency of the power grid; Total equivalent inertia of the system : Wherein: and are the number of nodes of the synchronous generator set and the new energy generator set respectively, and are respectively the inherent equivalent inertia corresponding to the th synchronous generator set and the virtual equivalent inertia corresponding to the th new energy generator set in each node, and are respectively the rated capacities of the th synchronous generator set and the th new energy generator set, is the total installed capacity of the system; Taking the new energy penetration rate and the total equivalent inertia of the system as inputs, construct a virtual equivalent inertia allocation optimization model, and iteratively adjust the virtual equivalent inertia weights and damping coefficients of each node through preset optimization constraints; Configure the optimized virtual equivalent inertia weights into the new energy grid-connected system model, inject a preset disturbance signal into the new energy grid-connected system model with the optimized virtual equivalent inertia weights configured, and perform dynamic simulation to generate dynamic response data; According to the dynamic response data, calculate the eigenvalues and damping ratios of the system's dominant oscillation mode through eigenvalue analysis, and quantitatively evaluate the small-signal stability index in combination with the spectral radius of the state matrix; Output the optimized virtual equivalent inertia weight configuration scheme and the stability assessment report. The optimized virtual equivalent inertia weight configuration scheme includes the recommended values of the virtual equivalent inertia of each node, the optimized damping coefficient, and the corresponding new energy penetration rate scenarios.

2. The small-signal stability simulation method for a new energy grid-connected system according to claim 1, wherein The step of taking the new energy penetration rate and the total equivalent inertia of the system as inputs, constructing a virtual equivalent inertia allocation optimization model, and iteratively adjusting the virtual equivalent inertia weights and damping coefficients of each node through preset optimization constraints includes: Objective function of virtual equivalent inertia distribution optimization model To minimize the two-norm of the system state matrix and the weighted frequency response robustness index of new energy penetration , the expression is as follows: Wherein: is the system state matrix constructed by the virtual equivalent inertia weight and the damping coefficient , is the new energy penetration rate which is defined as the ratio of the total capacity of new energy units to the total system capacity and are the weight coefficients is the frequency response correction factor is the frequency deviation to the disturbance frequency sensitivity matrix is the matrix infinity norm 3. The small-signal stability simulation method for the new energy grid-connected system according to claim 2, wherein The step of taking the new energy penetration rate and the total equivalent inertia of the system as inputs, constructing a virtual equivalent inertia allocation optimization model, and iteratively adjusting the virtual equivalent inertia weights and damping coefficients of each node through preset optimization constraints further includes: The constraint conditions of the virtual equivalent inertia allocation optimization model include node virtual equivalent inertia weight constraints, damping coefficient range constraints, and stability criterion threshold constraints. Among them, the stability criterion thresholds include the real part of the eigenvalue, the damping ratio, and the synchronizing torque coefficient.

4. The small-signal stability simulation method for a new energy grid-connected system according to claim 1, characterized in that, The step of configuring the optimized virtual equivalent inertia weights into the new energy grid-connected system model, injecting a preset disturbance signal into the new energy grid-connected system model with the optimized virtual equivalent inertia weights configured, and performing dynamic simulation to generate dynamic response data includes: Configure the optimized virtual equivalent inertia weights into the corresponding node controllers in the new energy grid-connected system model to update the inertia distribution parameters of the system; Inject a preset disturbance signal into the new energy grid-connected system model with the optimized virtual equivalent inertia weights configured, perform time-domain dynamic simulation, and record the frequency deviation curves of each node and the change curves of the power angle differences between synchronous units in real time to generate frequency response data and power angle oscillation characteristic data; From the frequency response data, calculate the difference between the maximum frequency and the steady-state frequency, and determine the frequency overshoot by dividing the difference by the steady-state frequency; and calculate the time from the start of the disturbance until the frequency returns within the steady-state error band as the settling time; extract the maximum instantaneous absolute value of the rotor angle difference of the synchronous generator unit from the rotor angle oscillation characteristic data as the maximum rotor angle swing.

5. The small-signal stability simulation method for a new energy grid-connected system according to claim 4, wherein Injecting a preset disturbance signal into the new energy grid-connected system model configured with the optimized virtual equivalent inertia weight includes: The preset disturbance signal includes at least one of a step load disturbance or a random power fluctuation disturbance; The step load disturbance is achieved by instantaneously increasing or decreasing the load at a specified bus, with the change amplitude being a preset percentage of the total system capacity and the duration being set as a short transient process; The random power fluctuation disturbance includes injecting a random power fluctuation sequence conforming to a statistical distribution at the connection point of the new energy generator units, and the fluctuation amplitude is determined based on the percentage of the rated capacity of the new energy generator units.

6. The small-signal stability simulation method for a new energy grid-connected system according to claim 4, wherein, Based on the dynamic response data, calculating the eigenvalues and damping ratios of the dominant oscillation modes of the system through eigenvalue analysis, and quantitatively evaluating the small-signal stability index in combination with the spectral radius of the state matrix, includes: Construct a state-space equation including the rotor dynamic equation of the synchronous generator unit, the virtual equivalent inertia control logic of the new energy generator units, and the power grid network equation based on the new energy grid-connected system model configured with the optimized virtual equivalent inertia weight; Calculate the eigenvalues of the state matrix in the state-space equation, and screen the low-frequency oscillation modes within a preset frequency range as the dominant oscillation modes; Calculate the damping ratio of the dominant oscillation mode, which reflects the attenuation speed of the oscillation amplitude; at the same time, calculate the spectral radius of the state matrix, which characterizes the maximum increase in the system dynamic response; If the damping ratio reaches or exceeds the preset lower limit and the spectral radius is less than or equal to the preset radius, it is determined that the small-signal stability of the system meets the standard; Otherwise, it is determined as not meeting the standard, and the virtual equivalent inertia weight and damping coefficient are re-optimized until the stability requirements are met.

7. The small-signal stability simulation method for a new energy grid-connected system according to claim 1, wherein Output the optimized virtual equivalent inertia weight configuration scheme and the stability evaluation report. The optimized virtual equivalent inertia weight configuration scheme includes the recommended values of the virtual equivalent inertia of each node, the optimized damping coefficient, and the corresponding new energy penetration scenarios, includes: Based on the optimized virtual equivalent inertia weight and damping coefficient, generate the recommended values of the virtual equivalent inertia of each node. The recommended values of the virtual equivalent inertia are bound to the new energy penetration scenarios, and the effectiveness is verified according to the frequency response data and rotor angle oscillation characteristic data in the dynamic simulation; Extract the damping ratio, spectral radius of the state matrix, and real part distribution of the eigenvalues of the dominant oscillation mode, and generate the stability level evaluation result in combination with the preset threshold; The stability evaluation report also includes the comparative analysis of the virtual equivalent inertia distribution under different penetration scenarios and the stability risk warning suggestions.

Images