Microgrid transient stability analysis method, device and equipment

Abstract

The invention provides a micro-grid transient stability analysis method, device and equipment, and relates to the technical field of transient analysis. The method comprises the following steps: updating a model parameter of each component in a target micro-grid based on operation data of the target micro-grid to obtain an updated model parameter of each component; constructing a system state space model of the target micro-grid according to the operation data of the target micro-grid and the updated model parameters of each component; and in the transient time window, determining the transient sensitivity of each component based on a preset disturbance quantity and a system state space model, and performing transient stability analysis on the target micro-grid by combining the operation data of the target micro-grid and the system state space model to obtain a transient stability analysis result of the target micro-grid. According to the method, the transient sensitivity of each component can be considered during transient stability analysis, so that the accuracy of transient analysis is improved, and the accuracy of a transient stability analysis result is ensured.

Description

Technical Field

[0001] This invention relates to the field of transient analysis technology, and in particular to a method, apparatus and equipment for transient stability analysis of microgrids. Background Technology

[0002] Microgrids, as a new type of power grid structure integrating distributed power sources, energy storage devices, loads, and monitoring and protection devices, rely on stable operation to ensure reliable power supply. Transient stability analysis of microgrids is an important means of assessing their ability to restore synchronous operation after large disturbances. Transient stability analysis typically relies on complex computer modeling and numerical simulation techniques. Because microgrids contain a large number of power electronic devices, exhibiting low inertia and weak damping characteristics, their dynamic behavior is complex and variable. Therefore, microgrids place higher demands on the data processing methods for transient stability analysis.

[0003] In existing technologies, transient stability analysis of microgrids mainly relies on time-domain simulations of all electromagnetic transients or electromechanical transients. Analysts typically build a microgrid simulation system containing detailed models of all components in specialized power system simulation software. After setting the initial operating conditions and disturbance events, they obtain the dynamic response curves of various electrical quantities of the system through long-term numerical integration calculations. Then, professionals judge the stability of the system based on the changing trends of the curves.

[0004] However, the component parameters used in existing technologies to build microgrid simulation models are mostly derived from typical values ​​in equipment manuals or general databases, which may deviate significantly from the actual operating conditions of microgrids. This can directly affect the accuracy of simulation results and may lead to misjudgments of microgrid stability. Summary of the Invention

[0005] This invention provides a method, apparatus, and device for transient stability analysis of microgrids, which solves the problem in the prior art that the impact of actual operating conditions on component parameters is not considered when constructing microgrid simulation models, resulting in insufficient accuracy of simulation results and thus affecting the stability judgment results of microgrids.

[0006] In a first aspect, embodiments of the present invention provide a method for transient stability analysis of a microgrid, comprising: The model parameters of each component in the target microgrid are updated based on the operating data of the target microgrid to obtain the updated model parameters of each component. Based on the operational data of the target microgrid and the updated model parameters of each component, a system state-space model of the target microgrid is constructed. Within the transient time window, based on the preset disturbance amount and the system state space model, the transient sensitivity index of each component is determined; Based on the transient sensitivity index of each component, the operating data of the target microgrid, and the system state-space model, a transient stability analysis of the target microgrid is performed, and the transient stability analysis results of the target microgrid are obtained.

[0007] Secondly, embodiments of the present invention provide a microgrid transient stability analysis device, comprising: The update module is used to update the model parameters of each component in the target microgrid based on the operating data of the target microgrid, so as to obtain the updated model parameters of each component. The module is used to construct the system state-space model of the target microgrid based on the operating data of the target microgrid and the updated model parameters of each component. The determination module is used to determine the transient sensitivity index of each component within the transient time window, based on the preset disturbance amount and the system state space model. The analysis module is used to perform transient stability analysis on the target microgrid based on the transient sensitivity index of each component, the operating data of the target microgrid, and the system state-space model, and obtain the transient stability analysis results of the target microgrid.

[0008] Thirdly, embodiments of the present invention provide an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the method described in the first aspect or any possible implementation thereof.

[0009] In this embodiment of the invention, the model parameters of each component in the target microgrid are updated using the operational data of the target microgrid to obtain updated model parameters. This ensures that the model parameters are more in line with the actual situation. By using the updated model parameters in combination with the operational data to construct a system state-space model, the system state-space model can more accurately reflect the dynamic characteristics of the microgrid. Within the transient time window, the transient sensitivity index of each component is determined using a preset disturbance sum system state-space model. By combining the transient sensitivity index, operational data, and system state-space model, transient analysis is performed on the target microgrid. This approach considers the transient sensitivity of each component during transient analysis, thereby improving the accuracy of transient stability analysis and ensuring the accuracy of the transient stability analysis report. Attached Figure Description

[0010] Figure 1 This is a flowchart illustrating the implementation of the microgrid transient stability analysis method provided in this embodiment of the invention. Figure 2 This is a flowchart illustrating the implementation of step S120 of the microgrid transient stability analysis method provided in this embodiment of the invention. Figure 3This is a flowchart illustrating the implementation of step S140 of the microgrid transient stability analysis method provided in this embodiment of the invention. Figure 4 This is a graph showing the dynamic response of the system frequency and the critical load bus voltage of the microgrid transient stability analysis method provided in this embodiment of the invention. Figure 5 This is a schematic diagram of the microgrid transient stability analysis device provided in an embodiment of the present invention; Figure 6 This is a schematic diagram of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0011] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0012] See Figure 1 The flowchart illustrating the implementation of the microgrid transient stability analysis method provided in this embodiment of the invention is described in detail below: Step S110: Update the model parameters of each component in the target microgrid based on the operating data of the target microgrid to obtain the updated model parameters of each component.

[0013] In some embodiments, the operational data of the target microgrid includes not only static system information of the target microgrid but also parameters required for dynamic models. Specifically, it is necessary to collect microgrid topology data to clarify the physical connections between various distributed power sources, energy storage devices, loads, and lines; collect line parameter data, i.e., the resistance, inductance, and capacitance values ​​of transmission and distribution lines; collect distributed power source model parameters, such as the phase-locked loop, proportional-integral parameters of inner and outer loop controllers, etc., for distributed power sources using inverter interfaces, and the inertia time constant, damping coefficient, and reactance of each axis, for synchronous generators; collect energy storage device model parameters, mainly involving the relevant control parameters of its charging and discharging control logic and energy management system; and collect load model parameters, such as ZIP model parameters for constant power, constant current, and constant impedance or more complex dynamic load model parameters.

[0014] It should be noted that when analyzing specific transient events such as load surges and short-circuit faults, the operational data should also include measured high-frequency time-series data that can accurately capture the dynamic response of the system. For example, when analyzing load surge scenarios, it is necessary to collect key electrical quantities within a short period before and after the disturbance, such as from 1 second before the disturbance to 5 seconds after the disturbance. Specifically, this may include: the effective value sequence of voltage and current at the node where the load experiencing the surge is located; power flow data on critical lines; dynamic curves of active and reactive power output of each distributed power source and energy storage unit; and real-time fluctuation data of the system frequency. These measured dynamic response data are the key basis for subsequent parameterization correction and effectiveness verification of the aforementioned theoretical model parameters, such as inverter controller parameters and load model parameters, thereby ensuring that the constructed system state-space model can accurately reproduce the transient behavior of the microgrid under real disturbances.

[0015] In some embodiments, model parameters obtained from equipment manuals or typical values ​​may deviate from actual operating conditions. Therefore, it is necessary to optimize the model parameters by extracting actual system responses, such as time-series data of key node voltages and branch currents obtained from actual measurements in the operating data, to obtain updated model parameters. This improves the accuracy of transient stability analysis of the target microgrid.

[0016] In one possible implementation, step S110 is specifically processed as follows: extracting the time series data of the target microgrid after disturbance from the operating data of the target microgrid, and determining the time series data as the actual system response; using multiple sets of candidate parameters of all components, performing time-domain simulation on the target microgrid to obtain multiple simulated system responses of the target microgrid; calculating the error between each simulated system response and the actual system response of the target microgrid according to a preset objective function; and determining the set of candidate parameters corresponding to the simulated system response with the smallest error as the updated model parameters of each component.

[0017] In some embodiments, the calculation process for the updated model parameters is as follows: First, it is necessary to extract time-series data containing dynamic processes from the operational data. Time-series data is not a steady-state snapshot, but rather a high-frequency sampling record of the voltage and branch current at key nodes after the system has experienced a small disturbance. Disturbances include planned load switching, capacitor bank switching, and natural fluctuations caused by background noise. The ideal data source is a synchronous phasor measurement unit (PMU), which can provide high-precision, timestamped voltage and current phasors. This time-series data is then recorded as the actual system response representing the actual measurements.

[0018] Secondly, system identification and optimization are performed to find an optimal set of parameters p from multiple candidate parameters. This optimal set of parameters ensures that the differential equation model based on this set of parameters, when subjected to the same perturbation excitation as in reality, produces a simulation output that is consistent with the actual situation. Compared with actual measurement data The error between them is minimized. Candidate parameters are obtained by making small changes to the model parameters, and each set of candidate parameters includes parameters from all components. The error is usually determined by an objective function. To quantify errors, the most common methods are root mean square error or weighted least squares. For example, the error quantization formula can be:

[0019] The summation iterates through each sampling point of the selected time series. , ||·|| represents the norm of a vector.

[0020] The iterative process is as follows: using a set of candidate parameters Initialize the differential equation model, apply the actual perturbation to the model, run a time-domain simulation, and obtain the simulated system response. Calculate the objective function The values ​​are used to generate a better set of candidate parameters based on the rules of the optimization algorithm, such as particle velocity and position updates in PSO. Repeat the aforementioned steps until the objective function is achieved. The algorithm converges to the minimum value or reaches the preset number of iterations. The optimal parameter vector obtained after convergence. This refers to the parameters after correction based on actual operating data.

[0021] Step S120: Construct the system state-space model of the target microgrid based on the operating data of the target microgrid and the updated model parameters of each component.

[0022] It should be noted that before constructing the system space model of the target microgrid, a precise parameter model of the target microgrid needs to be constructed first. Then, the system state space model of the target microgrid is constructed using this precise parameter model. When constructing the precise parameter model of the target microgrid, one can either first construct the parameter model using the model parameters, and then replace the model parameters in the parameter model with the updated model parameters to obtain the precise parameter model of the target microgrid; or one can directly construct the precise parameter model of the target microgrid using the updated model parameters.

[0023] See Figure 2 The specific processing method of the above step S120 includes steps S1201-S1203, and the specific content is as follows: Step S1201: Based on the operating data of the target microgrid and the updated model parameters of each component, construct the power flow equation of the target microgrid and the differential equation of each component.

[0024] Step S1202: Based on the differential equations of each component, determine the precise parameterized model of the target microgrid.

[0025] In some embodiments, when directly constructing an accurate parameterized model of the target microgrid using its operational data and updated model parameters for each component, it is necessary to establish an accurate parameterized model describing the dynamic behavior of the system based on the microgrid's topology and the physical relationships between its components. This model is a set of nonlinear differential-algebraic equations that characterizes the electromagnetic and electromechanical transient interactions between the components in the microgrid. For example, the inverter's control system is described by differential equations, while the power flow of the entire grid is constrained by algebraic equations. For instance, based on the topology and physical laws (such as Kirchhoff's laws), the dynamic characteristics of each component are described by differential equations such as the generator rotor motion equation and the inverter controller state equation, while the power balance of the entire grid is described by algebraic equations, i.e., the power flow equations, resulting in the power flow equations and the differential equations for each component. This constructs a set of nonlinear differential-algebraic equations describing the complete dynamic behavior of the system. The power flow equations are:

[0026] The differential equation for each component is:

[0027] in, This is a vector of state variables (such as power angle, frequency, and controller integral value). These are vectors of algebraic variables (such as node voltage magnitude and phase angle). It is a system parameter vector composed of the refined model parameters.

[0028] The differential equations of each component and the power flow equations of the target microgrid constitute the accurate parametric model of the target microgrid.

[0029] In some embodiments, when a parametric model of the target microgrid is first constructed using model parameters, and then the updated model parameters are used to replace the model parameters in the parametric model to obtain a more accurate parametric model of the target microgrid, it is necessary to establish a parametric model describing the dynamic behavior of the system based on the microgrid's topology, the physical relationships between its components, and the model parameters of each component, using these operational data. The parametric model includes the differential equations for each component and the power flow equations for the target microgrid. The power flow equations are as follows:

[0030] The differential equation for each component is:

[0031] in, This is a vector of state variables (such as power angle, frequency, and controller integral value). These are vectors of algebraic variables (such as node voltage magnitude and phase angle). This is the system parameter vector. Replace p with... The subsequent set of nonlinear differential-algebraic equations constitutes the exact parameterized model. Because its parameters are derived from the actual dynamic response, this model can more accurately reproduce the real dynamic behavior of the microgrid.

[0032] The specific process of constructing the differential equations and power flow equations is as follows: First, using the acquired microgrid topology data and line parameter data such as resistance and inductance, the network admittance matrix of the system is constructed using the nodal admittance method. This matrix constitutes an algebraic equation describing the power balance relationship of the entire network. At its core, it follows Kirchhoff's laws, which constrain the relationship between the voltage magnitude (a part of y) and phase angle (another part of y) of all nodes.

[0033] Secondly, while constructing the network admittance matrix, based on the parameters of the distributed power source model, the energy storage device model, and the load model, a differential equation is established for each dynamic component in the system. For example, for distributed power sources such as photovoltaics and energy storage connected to an inverter interface, the differential equations primarily describe the dynamics of their control system. The x-vector contains integrator state variables from the phase-locked loop (PLL), and integrator state variables from the PI controller in the internal or external current loop or power voltage loop. The specific form of the function f is determined by the control strategy, such as voltage source control or current source control. For synchronous generators, the differential equations mainly consist of the swing equation describing rotor motion and the state equations describing the excitation system and speed control system. In this case, the x-vector contains power angle, speed deviation, and excitation winding flux linkage. For dynamic loads, such as large induction motors, the differential equations describe rotor slip and circuit dynamics. The initial nonlinear differential-algebraic equations obtained in this case, with system parameter vector p such as the PI controller gain of the inverter and the inertia time constant of the synchronous generator, mainly originate from equipment manuals or industry typical values, which form the basis for subsequent parameter correction.

[0034] Step S1203: Determine the system state-space model of the target microgrid based on the precise parameterized model and power flow equations.

[0035] In some embodiments, since the precise parameterization model is a nonlinear model, and nonlinear models are extremely complex when performing sensitivity analysis and large-scale calculations, it is necessary to linearize the precise parameterization model at the current system operating point (i.e., the precise equilibrium point) to obtain the system state-space model of the target microgrid.

[0036] In one possible implementation, step S1203 is specifically processed as follows: using the precise parameterized model, the power flow equation is solved to obtain the precise equilibrium point of the target microgrid; at the precise equilibrium point, the precise parameterized model is linearized to obtain the system state-space model of the target microgrid.

[0037] In some embodiments, the precise equilibrium point of the system needs to be determined by solving the power flow equations of the target microgrid. This precise equilibrium point, also known as the operating point, is then used to expand the nonlinear differential-algebraic equations (i.e., the differential equations for each component, including the updated model parameters for each component) at this point using Taylor series expansion, neglecting higher-order terms, to obtain a linear system state-space model. This model is typically expressed in the following form:

[0038] in, It is the state deviation vector of the system state variables relative to the operating point, such as generator power angle deviation, inverter DC side voltage deviation, etc. It is the derivative of the state variable deviation with respect to time; It is a vector of deviations of the input variable relative to the operating point, such as a small change in load; It is the system state matrix, whose elements are determined by the partial derivatives of the system at the operating point, reflecting the coupling relationship between the internal state variables of the system; This is the input matrix, reflecting the impact of input changes on the system state. This equation is the state-space equation of the target microgrid.

[0039] In some embodiments, by using a nonlinear system of differential-algebraic equations in Performing a Taylor series expansion on the points and ignoring higher-order terms (second order and above), we obtain a linear system model. and At the running point After linearization, we get:

[0040]

[0041] in , , is the deviation of state variables and algebraic variables relative to the running point. , , , These are the Jacobian matrices obtained by taking the partial derivatives of functions f and g with respect to variables x and y, respectively, and they are at the running point. Perform the evaluation. This represents a small external perturbation input. and This is the corresponding input matrix. To obtain the standard state-space form... It is necessary to eliminate algebraic variables. From the second linearized equation, we can obtain:

[0042] Substituting this equation into the first linearized equation and rearranging, we obtain the final state-space model of the linear system:

[0043] Therefore, the final system state matrix A and input matrix B are:

[0044]

[0045] Finally, the system state-space model is generated. The resulting system state-space model represents the dynamic characteristics of the microgrid under specific operating conditions in a linear manner, and serves as the mathematical basis for subsequent transient sensitivity analysis and stability assessment.

[0046] Step S130: Within the transient time window, determine the transient sensitivity index of each component based on the preset disturbance amount and the system state space model.

[0047] In some embodiments, the transient sensitivity index refers to the degree of response and tolerance of the operating state, control performance, and stability of components such as photovoltaic inverters, energy storage converters, and diesel generators in the target microgrid to transient disturbances such as voltage sags and short-circuit faults. This index exhibits differentiated characteristics due to variations in equipment structure and control strategies. Accurately assessing this index is crucial for ensuring the transient stability of the microgrid and for optimizing its configuration and control. The transient sensitivity index of each component can be obtained by simulating these disturbances by setting preset disturbance values ​​and further simulating them using a system state-space model.

[0048] In one possible implementation, step S130 is specifically processed as follows: a preset perturbation amount is applied to the updated model parameters of each component to obtain the perturbed system state space model; partial derivatives are calculated on both sides of each state space equation in the perturbed system state space model to obtain the state trajectory sensitivity equation of each component; the state trajectory sensitivity equation of each component is solved using numerical integration and the system state space model to obtain the sensitivity numerical trajectory of each component; for each component, the sensitivity of the component at each moment within the transient time window is extracted from the sensitivity numerical trajectory of the component, and the sensitivity at each moment within the transient time window is weighted and integrated to obtain the transient sensitivity index of the component.

[0049] In some embodiments, the transient sensitivity index is calculated by utilizing the obtained system state-space model to establish a corresponding state trajectory sensitivity equation for each component parameter to be analyzed in the microgrid. The state trajectory sensitivity is the partial derivative of the system state variable with respect to the refined model parameters of a specific component; it describes the degree of deviation of the system state trajectory when the parameter undergoes a small change. For any component's refined model parameters... Its state trajectory sensitivity equation can be obtained by simultaneously finding the equation about both sides of the original state space equation. The partial derivatives yield:

[0050] in, With accurate model parameters If it is irrelevant, then we can conclude that:

[0051] at this time, It can be represented as Therefore, refining the model parameters The state trajectory sensitivity equation can be expressed as:

[0052] in, This reflects the influence of system state variables on the parameters of the accurate model. How quickly the sensitivity changes over time, It represents the state trajectory to the refined model parameters The sensitivity value, its initial value It is usually set as the zero vector, indicating that the system is in a steady state before the disturbance occurs, and the sensitivity of the state to the parameters is zero; It is the system state matrix, which was obtained during the model building phase; It is the state trajectory of the system under a specific disturbance, obtained by performing time-domain simulation on the original state-space model; It is a state matrix For accurate model parameters The partial derivative matrix, the calculation of which requires tracing back to the physical differential equations of the system model and analyzing the parameters of the refined model. How does it affect the state matrix? Find the elements in the set and calculate their corresponding analytic derivatives.

[0053] In some embodiments, at a determined operating point, based on The reference state matrix A is calculated. ),right Apply a small perturbation Δp, for example, Δp is The value is reduced by 0.1%, resulting in a new parameter value. '= +Δp, keeping other system parameters and operating point unchanged, relinearize the system model to obtain the perturbed state matrix A( '). Each element in The following central difference formula can be used for approximate calculation:

[0054] Alternatively, one-sided difference can be used:

[0055] By repeating this process for each refined model parameter to be analyzed, the required partial derivative matrix can be obtained systematically and automatically. .

[0056] By simultaneously solving the state trajectory sensitivity equation with the original system's state-space model using numerical integration methods such as the Runge-Kutta method, the sensitivity of each state variable to the system's state space during the entire dynamic process can be obtained. The trajectory of sensitivity value changing over time The sensitivity numerical trajectory in the time domain is obtained. Subsequently, to obtain a scalar index that can comprehensively evaluate the influence of this parameter throughout the entire transient process, it is necessary to perform a weighted integral of the sensitivity value within a preset transient time window. This transient time window, for example, starts from the moment the fault occurs. At a certain assessment endpoint after fault clearance This covers the key stages of the system's dynamic response. The integration process is completed using the following formula, thereby generating the final transient sensitivity index. :

[0057] in, yes The corresponding transient sensitivity index; It is a preset transient time window; It is a sensitivity vector obtained by solving the state trajectory sensitivity equation; Represents the sensitivity vector The square of the Euclidean norm combines the sensitivity of all state variables into a single, always positive value, highlighting the sensitivity components with larger deviations. It is a non-negative weighting function used to adjust the importance of sensitivity at different times. For example, higher weight can be given to the peak time when the system oscillates. If no special consideration is given, it can be set to a constant of 1. Repeating the above process for all the key component parameters to be analyzed in the microgrid will yield a series of transient sensitivity indices, which together form the basis for evaluating the influence of each component.

[0058] The following concrete example further illustrates this process: Assume the change in the operating parameter of interest is the system frequency. The dynamic response, particularly its maximum drop depth and oscillation amplitude during recovery, is assessed. The aim is to evaluate the virtual inertia of a specific energy storage inverter in a microgrid. The extent of the impact on this frequency response process, As parameters to be analyzed State vector This must include a state variable representing the system frequency deviation, denoted as . Therefore, the obtained sensitivity vector In, the corresponding component The physical meaning is that at time t, if the virtual inertia coefficient A tiny increase of one unit results in a change in system frequency deviation. How much will change accordingly? The trajectory curve reveals the parameters The magnitude and direction of the influence on frequency deviation at different stages of the transient process. Weighting function. This is key to focusing on changes in specific characteristics. To specifically assess the impact on frequency drops and oscillations, It should not be a constant of 1, but rather designed as a function related to the frequency deviation itself. An effective choice is: ,use The physical meaning is that the further the system frequency deviates from the rated value, the more dangerous the system is considered to be. Sensitivity at this moment The more important a parameter is, the greater its weight should be. This ensures that the final sensitivity metric more accurately reflects the parameter's influence on the system's most vulnerable moments. In this scenario, the integration process is completed using the following formula, thus generating the final transient sensitivity, denoted as... :

[0059] in, It is for parameters In this example, it is The transient sensitivity index. The subscript here... This is a definitive label that clearly indicates that the indicator focuses on evaluating frequency dynamics. It has an impact. It is a preset transient time window that covers the period from the moment the fault occurs. When the system frequency response is basically stable . The absolute value of the system frequency deviation trajectory obtained from the simulation of the original state-space model is used as the dynamic weight. Still representing the entire sensitivity vector The square of the Euclidean norm. This means that the parameters are still considered. The combined effect on all state variables is considered, but the importance of this effect is modulated by the degree of deviation from the current frequency. The resulting transient sensitivity index... It is a specific numerical value, referring to frequency dynamics. Transient sensitivity. This is achieved through dynamic weighting, measuring specific component parameters such as energy storage virtual inertia. A comprehensive quantitative indicator of the impact of the system's overall state trajectory deviation throughout the transient process, especially during periods of drastic frequency fluctuations. The generated indicator can be labeled as... That is, the virtual inertia of the energy storage unit The transient sensitivity index. If calculated Another parameter, such as the gain of a certain control loop in a photovoltaic inverter, is... Indicators The result clearly indicates that in this frequency transient event caused by the short-circuit fault, the virtual inertia of the energy storage has a far greater effect on stabilizing the system frequency dynamics than the controller gain of the photovoltaic inverter. A larger value indicates a stronger amplification / suppression capability of the component parameter on system frequency disturbances during transient events, making it a key parameter affecting system frequency stability. By repeating the above process for all key component parameters to be analyzed in the microgrid, a series of transient sensitivity indices with clear physical labels, such as TFSI representing frequency sensitivity and TVSI representing voltage sensitivity, can be obtained.

[0060] Step S140: Based on the transient sensitivity of each component, the operating data of the target microgrid, and the system state-space model, perform transient stability analysis on the target microgrid to obtain the transient stability analysis results of the target microgrid.

[0061] In some embodiments, when performing transient stability analysis, it is necessary to first identify multiple key components from the components, and then combine the key components to perform transient stability analysis on the target microgrid. The transient stability analysis results of the target microgrid include the transient stability margin of the target microgrid and system optimization suggestions, and may also include other contents, without specific limitations.

[0062] See Figure 3 The specific processing method of step S140 above includes steps S1401-S1407, the details of which are as follows: Step S1401: Based on the transient sensitivity index of each component and the system state-space model, select several key components from all components of the target microgrid.

[0063] In some embodiments, before screening key components, a sensitivity threshold needs to be set to distinguish the degree of influence. The sensitivity threshold can be set based on engineering experience, historical data analysis of benchmark systems, or statistical methods. For example, the values ​​corresponding to the top 20% of the transient sensitivity indices of all components can be selected as the threshold. The aim is to initially identify components that have a significant impact on the transient response of the system. Then, in combination with this sensitivity threshold, multiple key components can be screened from all components of the target microgrid.

[0064] In one possible implementation, step S1401 is specifically processed as follows: the component corresponding to the transient sensitivity index that exceeds the sensitivity threshold is identified as a candidate key component; the value corresponding to the off-diagonal element in the system state matrix of the system state space model is identified as the coupling strength of the two components corresponding to the off-diagonal element; when the coupling strength between the first component and the second component exceeds a preset coupling threshold, and the first component is a candidate key component while the second component is not a candidate key component, the second component is identified as a coupling key component; all candidate key components and coupling key components are identified as key components.

[0065] In some embodiments, by comparing the transient sensitivity index of each component with a sensitivity threshold one by one, components whose transient sensitivity index exceeds the threshold can be initially screened as candidate key components. However, relying solely on a single sensitivity index threshold may overlook the interactions between components. Therefore, it is necessary to analyze the dynamic coupling relationships between these candidate key components in depth based on the system state-space model, and further screen coupled key components from the candidate key components based on the dynamic coupling relationships between different components. Finally, both candidate key components and coupled key components are determined as key components. The system state matrix in the system state-space model can reflect the dynamic coupling relationships between different components. The preset coupling threshold is set in advance; different components can use the same preset coupling threshold, or different preset coupling thresholds can be set for different components based on their characteristics.

[0066] It should be noted that dynamic coupling describes the degree to which the state change of one component affects the state change of another component. In the system state-space model, the system state matrix... off-diagonal elements Directly reflects the state variables For state variables The impact of the rate of change. If two different candidate key components have core state variables in the state matrix... If the absolute value of the corresponding off-diagonal element is large, it indicates a strong dynamic coupling between the two components. This can be achieved by analyzing the state matrix. The structure and values, especially the submatrices associated with candidate key components, can quantify and identify these close dynamic interactions.

[0067] By combining sensitivity thresholds with the coupling strength between different components to identify key components, a comprehensive approach can be taken into account both the sensitivity of individual components and the coupling strength between them. The criteria for a component to be ultimately identified as a key transient impact component are: first, its own transient sensitivity index is significantly higher than the set sensitivity threshold, indicating that it is an independent dominant transient factor; second, if its transient sensitivity index is close to or slightly higher than the threshold, it has a significant dynamic coupling relationship with one or more components that have been identified as key. This verification considering coupling relationships can avoid overlooking components with moderate individual influence but which significantly impact the system's transient processes through strong coupling with other key components.

[0068] For example, when performing transient analysis on a microgrid that includes a photovoltaic (PV) power plant, an energy storage system (ESS), a synchronous generator (SG), and multiple industrial loads, the following steps are required: The first step is to set a sensitivity threshold and screen candidate key components. The specific objective of this analysis is to evaluate the system's frequency stability in the event of a three-phase short-circuit fault in a nearby line. Therefore, the transient sensitivity index (TFSI) to system frequency deviation is first calculated for key parameters of each component, such as the phase-locked loop bandwidth of the PV inverter, the virtual inertia coefficient of the ESS, and the governor droop coefficient of the SG. A set of values ​​is obtained after calculation; for example, the TFSI of the ESS is 0.95, the TFSI of the SG is 0.88, the TFSI of a large PV power plant is 0.65, while the TFSI of other small distributed power sources and loads is below 0.2. The threshold is set based on the statistical method of "selecting the values ​​corresponding to the top 20% of the transient sensitivity indices of all components after sorting." Assuming the lowest value of the top 20% after sorting is 0.80, this 0.80 is determined as the sensitivity threshold. Subsequently, the TFSI of all components was compared with 0.80. The ESS was 0.95 > 0.80 and the SG was 0.88 > 0.80, both exceeding the threshold. Therefore, they were initially screened out to form an initial list of candidate key components: {energy storage system, synchronous generator}.

[0069] The second step involves analyzing the dynamic coupling relationships between components using a system state-space model and examining other components with sensitivity close to the threshold. In the microgrid state-space model, state variables are assumed... The speed deviation of the synchronous generator SG is represented by the state variable x. 15 This represents the state of the power controller integrator used for frequency support in the ESS inverter of the energy storage system. Check the corresponding off-diagonal elements in the system state matrix A. If the calculation reveals its absolute value This value is among the highest in the off-diagonal elements of the entire matrix, directly and quantitatively demonstrating that changes in the power control state of the energy storage system have an extremely strong and rapid impact on the rate of change of the synchronous generator's speed. Similarly, The value is also quite large, indicating a close two-way dynamic coupling between the two. Meanwhile, it's noted that the TFSI of the large PV power plant is 0.65, which, although not reaching the threshold of 0.80, is not low. Further analysis of its core state variables, such as... This represents the state of its power voltage control loop, and is related to SG. The coupling between them reveals the corresponding off-diagonal elements. The coefficient of performance is only 2.1, far less than the aforementioned coupling strength. This indicates that, under the current operating conditions, the dynamic interaction between large PV power plants and the core support source (SG) for system frequency stability is relatively weak.

[0070] The third step involves comprehensive verification to determine the final set of critical transient impact components. Based on the results of the previous analysis, the final screening criteria are applied. First, the transient sensitivity index (TFSI) of the energy storage system (ESS) is 0.95, and that of the synchronous generator (SG) is 0.88, both exceeding the set sensitivity threshold of 0.80. According to criterion one, they are independent dominant transient factors and are therefore directly identified as critical transient impact components. Next, the large PV power plant is examined, with a TFSI of 0.65, close to but below the threshold. According to criterion two, it is necessary to check whether it has significant dynamic coupling with the components already identified as critical. In the coupling analysis in the previous step, the dynamic coupling strength between the PV power plant and the SG was found to be |A7, 20 The coefficient |=2.1, which does not exceed the preset coupling threshold. Therefore, the large PV power plant does not meet criterion two. Although it has some influence, it is neither an independent strong influence source nor a member of a strongly coupled group, and therefore is not included in the final set. Through this verification, the final set of key transient influence components is determined to be {energy storage system, synchronous generator}. The physical meaning of this set is that during the short-circuit fault transient process of this microgrid, the hybrid inertia support system composed of energy storage and synchronous generator is the core for maintaining frequency stability. Subsequent model reduction and simulation analysis should focus on retaining the detailed dynamic models of these two components.

[0071] Step S1402: Based on the target disturbance scenario and multiple key components, the simulation model of the target microgrid is reduced in order to obtain the dynamic equivalent model of the target microgrid for disturbance analysis.

[0072] It should be noted that during transient processes, not all components have an equally significant impact on the overall dynamic response of the system. Therefore, this method divides the target microgrid into two parts: a dynamic core region requiring detailed modeling and a simplified region that can be reduced in order. By reducing the order of the simplified region in the simulation model of the target microgrid, a dynamic equivalent model for disturbance analysis can be obtained. This dynamic equivalent model can accurately reflect the core dynamic response of the system under specific disturbances while being computationally efficient. The division of the target microgrid needs to be completed in conjunction with the target disturbance scenario and multiple key components.

[0073] The disturbances in the target disturbance scenarios are predefined typical operating conditions used to assess the transient stability of microgrids. These typically include three types that pose a severe test to system stability: load abrupt changes, such as the starting of large induction motors or the disconnection of critical loads; renewable energy power fluctuations, such as a sharp drop in power from photovoltaic arrays due to cloud cover or a sudden surge in power from wind turbines due to gusts; and short-circuit faults, especially severe symmetrical faults such as three-phase short circuits. Each target disturbance scenario requires clearly defined key parameters such as its type, location, intensity, and duration. After identifying the specific disturbance scenario, the next step is to define the range of model order reduction based on the type and location of the scenario, combined with the set of key transient impact components identified in the previous step.

[0074] In one possible implementation, step S1402 is specifically processed as follows: the region where the electrical distance from each disturbance point in the target disturbance scenario exceeds a preset distance is identified as a reduced-order region; all components that are not critical components in the reduced-order region are simplified equivalently to obtain the dynamic equivalent model of the disturbance analysis of the target microgrid.

[0075] In some embodiments, critical components form the basis of the dynamic core region of the target microgrid, and their detailed dynamic models must be fully preserved regardless of their physical location. Furthermore, components within the disturbance initiation point and its electrical proximity, whose dynamic behavior is crucial due to direct exposure to disturbance impacts, must also be included in the detailed modeling region. Therefore, the reduced-order region is defined as all areas in the microgrid that are not part of the critical transient impact component set and are electrically distant from the disturbance point. Finally, based on the determined critical transient impact component set and the reduced-order region, equivalent simplification is performed on non-critical components within the reduced-order region to generate the final dynamic equivalent model for disturbance analysis. Equivalent simplification utilizes model reduction techniques to replace the dynamic behavior of a complex set of non-critical components with one or a few simpler equivalent models, while ensuring that the equivalent model can reproduce the external characteristics of the original system at its boundary points connecting to the rest of the system.

[0076] It should be noted that commonly used model reduction methods include: based on the coherence criterion of electromechanical oscillation modes, aggregating distributed sources with consistent oscillation behavior in the transient response into an equivalent source; and using modal analysis to identify and eliminate dynamic modes in the external system that have little impact on the internal system, decay rapidly, or have frequencies much higher than the research range. By performing this processing on all non-critical components within the reduction region, the final dynamic equivalent model for disturbance analysis is a hybrid model. It retains a high-precision differential equation model for critical components, while using a low-order equivalent model with significantly reduced computational complexity for the rest of the system.

[0077] Taking a microgrid system containing three generators G1, G2, and G3 as an example, the specific process of order reduction is as follows: First, the disturbance scenario is a three-phase short-circuit fault occurring at the end of line 5-7 near bus 7, with the fault occurring at 1.0 second and clearing at 1.15 seconds; it is assumed that G1 has been identified as a key transient impact component through transient sensitivity analysis.

[0078] Secondly, based on the location of the disturbance (located on bus 7), the transfer impedance from the fault point to buses 1, 2, and 3 where each generator is located is calculated in the node impedance matrix, thus obtaining... .、 .、 The impedance threshold for model order reduction is set to 0.2 pu. Since... Therefore, the region surrounding G3 is determined as the area for model order reduction; at the same time, although However, G2 is not a critical transient component and its physical location is close to G3, so it is usually included in the consideration. In this area, it is necessary to determine the coherence of non-critical components G2 and G3. By performing a short-time simulation on the original system, it was observed that the rotor angle difference between G2 and G3 during the disturbance does not exceed the preset difference threshold of 5 degrees, that is, the oscillation behavior is consistent.

[0079] Finally, G2 and G3 are simplified by aggregation, assuming that the inertial time constant of G2 is... The baseline capacity is The inertial time constant of G3 is The baseline capacity is Set the baseline capacity of the equivalent unit. The equivalent inertial time constant is calculated based on the formula. A dynamic equivalent model for disturbance analysis was obtained. This model retains the detailed model of the key component G1, while G2 and G3 are represented by a device with an inertial time constant of 4.50s and a reference capacity of [missing information]. The equivalent generator is replaced by this model, which will be used for subsequent efficient time-domain simulation analysis.

[0080] Step S1403: Control the dynamic equivalent model of disturbance analysis of the target microgrid, and simulate it according to the operating data of the target microgrid to obtain the time-domain simulation data of the target microgrid.

[0081] In some embodiments, after constructing a dynamic equivalent model for disturbance analysis, the dynamic response of a microgrid under a specific disturbance scenario can be simulated by numerical calculation to obtain time-domain simulation data for subsequent stability assessment.

[0082] The specific process for acquiring time-domain simulation data is as follows: First, using the operational data obtained in the previous steps, precise initial simulation conditions are set for the dynamic equivalent model of the disturbance analysis. These initial conditions are the steady-state operating point of the system before the disturbance occurs, including the magnitude and phase angle of the voltage at each node, the active and reactive power flowing through each branch, and the initial values ​​of the state variables of all dynamic components in the model, such as the initial value of the integrator in the inverter control loop, and the initial power angle and speed of the synchronous generator. By performing power flow calculations on the system before the disturbance, the initial values ​​of these algebraic and state variables can be accurately solved, ensuring that the simulation starts from a realistic, physically consistent system equilibrium point.

[0083] Secondly, numerical integration is employed to solve the dynamic response of the dynamic equivalent model of perturbation analysis under the selected perturbation scenario. The dynamic equivalent model of perturbation analysis is essentially a system of differential-algebraic equations, which cannot be solved analytically; therefore, numerical methods must be used to solve it.

[0084] It should be noted that commonly used numerical integration methods include explicit methods such as the improved Euler method and the fourth-order Runge-Kutta method, or implicit methods such as the trapezoidal integration method. The choice of method depends on the rigidity of the model and the required computational accuracy.

[0085] The simulation process begins at the moment the disturbance occurs and progresses in small time steps. Within each time step, the numerical integration algorithm calculates the state variable values ​​for the next moment based on the current state variable values ​​and the system equations. For example, if the disturbance scenario involves a three-phase short circuit on a line at t=0.1 seconds and the fault being cleared at t=0.2 seconds, the simulation software will simulate this event sequence and calculate the complete evolution of the system state over time. Throughout the numerical integration process, key dynamic response data needs to be recorded in real time.

[0086] At each simulation time step, physical quantities that directly reflect the transient stability of the microgrid, such as voltage amplitude, system frequency, or power angle of key generators, are recorded for pre-specified key nodes or all nodes. For example, the voltage change curve of key load nodes over time is recorded to observe whether voltage collapse occurs; the fluctuation curve of system frequency is recorded to determine whether frequency instability occurs; and the relative differences in power angles among key generators are recorded to assess power angle stability. These recorded, time-series data of voltage, frequency, and power angle are then organized to generate the time-domain simulation data for subsequent analysis.

[0087] Step S1404: When the dynamic response curve in the time domain simulation data converges, the equivalent model of control disturbance analysis is used for iterative simulation to obtain the transient stability margin of the target microgrid.

[0088] In some embodiments, if the dynamic response curve in the time-domain simulation data eventually converges to a new stable operating point after a period of oscillation, the target microgrid is determined to be transiently stable; conversely, if the dynamic response curve shows continuous divergence or collapse, the target microgrid is determined to be in an unstable state. It is necessary to extract key state variables from the time-domain simulation data, such as the voltage amplitude of key nodes, system frequency, and relative power angle between key generators, and plot the dynamic response curve.

[0089] It should be noted that when the dynamic response curve converges, that is, when the target microgrid recovers stability under the current disturbance, it is necessary to further quantify its distance from the instability boundary, i.e., calculate the transient stability margin. The calculation process for the stability margin is as follows: First, obtain the current running state, which is the baseline operating condition for the initial time-domain simulation.

[0090] Secondly, the stability boundary of the system is found through a series of iterative simulations. This process typically employs a critical parameter search method. For example, if the disturbance is a short-circuit fault, the duration of the fault, i.e., the critical clearing time, can be gradually increased; if the disturbance is a sudden load increase, the power of the increased load can be gradually increased. In each iteration, the time-domain simulation is repeated using new disturbance parameters, and the stability of the system is determined. In this way, a disturbance parameter value that places the system in a critical state between stability and instability can be found. The stability boundary of the system is defined by this critical disturbance parameter.

[0091] Finally, a quantified transient stability margin is generated by calculating the difference between the critical disturbance parameter corresponding to the stability boundary and the actual disturbance parameter under the current operating state. For example, the transient stability margin can be expressed as the difference between the critical clearing time and the actual protection action time, or the maximum load increment that the system can withstand.

[0092] Step S1405: Perform correlation analysis on the transient stability margin of the target microgrid and multiple key components to obtain the correlation analysis results.

[0093] In some embodiments, after obtaining the quantified value of the transient stability margin, it is necessary to perform correlation analysis with the previously selected set of key transient impact components. Through correlation analysis, the characteristics or operating states of the key components that play a decisive role in the magnitude of the transient stability margin can be determined. For example, it is possible to analyze how the transient stability margin changes after changing the control parameter of a key distributed power source, thereby determining the optimization direction of that parameter. By systematically studying the interrelationships between each element in the set of key transient impact components and the transient stability margin, a series of correlation analysis results can be obtained.

[0094] Step S1406: Based on the correlation analysis results, system optimization suggestions for the target microgrid are obtained.

[0095] In some embodiments, the correlation analysis results can be used to further generate system optimization recommendations for improving system stability. These recommendations are specific and actionable; for example, they may suggest adjusting the power response rate of a key energy storage device, optimizing the voltage control strategy parameters of a key inverter, or adding additional dynamic reactive power compensation equipment at a key location in the microgrid.

[0096] Step S1407: The transient stability margin and system optimization suggestions of the target microgrid are determined as the transient stability analysis results of the target microgrid.

[0097] In some embodiments, by integrating all analysis results, a comprehensive transient stability analysis can be generated. The transient stability analysis results include a clear quantitative assessment of the transient stability margin under the current microgrid operating conditions, indicating the system's safety redundancy in response to specific disturbances; it also includes the key components that have the greatest impact on system stability, as well as targeted system optimization suggestions that can improve transient stability margins using correlation analysis results.

[0098] The following example illustrates the specific process of obtaining correlation analysis results using transient stability margin, deriving system optimization suggestions based on the correlation analysis results, and finally obtaining the transient stability analysis results of the target microgrid: Taking an industrial park microgrid comprising a photovoltaic power plant, wind turbines, a Bus Estimated Energy Storage System (BESS), and a Synchronous Generator (SG) as an example, the pre-defined disturbance scenario is a 120ms three-phase short-circuit fault occurring on the critical feeder L-5 within the park, and the set of key transient impact components has been identified through previous steps as {BESS, SG}. The correlation analysis is specifically implemented by calculating the normalized sensitivity of the Transient Stability Margin (TSM) to the core parameters of the key components in this set.

[0099] At this point, the transient stability margin is quantified using the critical resection time (CCT), and the actual protective action time is... Then the transient stability margin The transient stability margin can be viewed as an implicit function of the system parameter p. The normalized sensitivity is calculated, defined as the ratio of the change in stability margin to the relative change in the parameter. This sensitivity is calculated using the following detailed numerical method: The first step is to calculate the baseline transient stability margin. Based on the constructed reduced-order dynamic equivalent model of perturbation analysis, all system parameters are set as baseline values. A bisection search is used to iteratively find the critical cut-off time within the possible interval [120ms, 500ms]. For example, if the initial test time is set to (120+500) / 2 = 310ms, and the system becomes unstable after one time-domain simulation, the upper bound is updated to 310ms. If the system stabilizes after another test at (120+310) / 2 = 215ms, the lower bound is updated to 215ms. This process is repeated until the interval width is less than a preset precision, such as 1ms. Assume the final calculated baseline critical cut-off time is... Therefore, the baseline transient stability margin is: .

[0100] The second step involves perturbing the parameters of key components one by one and calculating their sensitivity. First, for the parameters of the BESS (Body Energy Storage System), its virtual inertia is selected. As parameters to be analyzed Benchmark value Apply a small positive perturbation The new parameter value is Keeping all other parameters in the model unchanged, only... Updated to Rerun the bisection search to calculate the new critical resection time, assuming it is... The formula for calculating sensitivity is:

[0101] This result indicates that For every 1% increase in parameters, the transient stability margin CCT will increase by 1.8 ms.

[0102] Secondly, based on the parameters of the synchronous generator SG, the proportional gain of the automatic voltage regulator (AVR) in its excitation system is selected. As parameters to be analyzed Benchmark value Apply a small positive perturbation The new parameter value is Keep all other parameters in the model at baseline values, and only set the baseline values. Updated to Recalculate the critical resection time, assuming it is... The formula for calculating sensitivity is:

[0103] This result indicates that For every 1% increase in parameters, the transient stability margin CCT decreases by 0.6 ms, showing a negative correlation. This is because excessively high gain exacerbates the second pendulum or subsequent oscillations of the power angle under this perturbation, which is detrimental to transient stability.

[0104] The third step is to generate a sensitivity list and formulate system optimization suggestions. The second step is repeated for all pre-selected core parameters in the set of critical transient impact components, such as the droop coefficient of the BESS and the damping winding parameters of the SG, to obtain a complete sensitivity list, as shown in Table 1, and sorted by the absolute value of the sensitivity.

[0105] Table 1 Sensitivity List

[0106] Based on this sensitivity list, by comparing the absolute values ​​of the normalized sensitivity (|+180|>|-60|>|+30|), the optimization priority of the parameters can be clearly and objectively determined, generating the following system optimization suggestions: For the virtual inertia of the BESS energy storage system... This parameter has the highest positive normalization sensitivity of +180ms. It is recommended to "increase the virtual inertia of the energy storage inverter." Increasing the transient stability margin (CCT) by 10% from 3.5s to 3.85s is expected to improve the transient stability margin (CCT) by approximately 180ms * 10% = 18ms. This is the most effective way to improve the transient stability of the system. (This is related to the AVR gain of the synchronous generator SG.) This parameter has the second-highest negative normalization sensitivity in absolute value: -60ms. It is recommended to "avoid further increasing the current AVR gain." If steady-state voltage regulation allows, you can try... Reducing the transient stability margin by 10% from 200 p.u. to 180 p.u. is expected to improve the transient stability margin by approximately |-60 ms| * 10% = 6 ms, and this needs to be verified through simulation. Regarding the qv droop coefficient for the BESS energy storage system... This parameter has low sensitivity to normalization. Optimizing this parameter yields limited benefits, but it can still be considered as an alternative.

[0107] Baseline transient stability margin The quantitative evaluation results are integrated with the system optimization suggestions based on normalized sensitivity analysis and ranked by priority. The final transient stability analysis results not only clearly indicate safety redundancy, but also provide the most cost-effective parameter optimization path obtained through quantitative comparison.

[0108] The transient stability analysis results include the baseline transient stability margin calculated under specific disturbance scenarios, such as... This is used to quantify the system's safety redundancy; it includes a set of critical components most affected by transient processes, such as the energy storage system BESS and the synchronous generator SG, and presents a detailed normalized sensitivity list as the basis for decision-making. Based on this list, it provides system optimization recommendations ranked by priority. For example, the primary recommendation is to adjust the energy storage virtual inertia, which has the highest positive sensitivity. The expected margin for improvement is included, such as an 18ms increase. The second adjustment involves adjusting the AVR gain, which has the second-highest absolute value sensitivity. It should also include necessary risk warnings and verification requirements; finally, it can also include dynamic response curves for key variables.

[0109] To verify the feasibility of this invention in practice, it was applied to a microgrid in an industrial park. This industrial park microgrid aims to achieve energy self-sufficiency and high-reliability power supply, but it contains a large number of distributed photovoltaic and wind turbine generators, as well as impactful industrial loads. Power fluctuations from renewable energy sources and the start-up and shutdown of large motors pose a severe challenge to the system's transient stability. Traditional full-scale time-domain simulation methods are computationally intensive and time-consuming, making it difficult to meet the timeliness requirements for operational assessment and contingency plan development for this microgrid.

[0110] In this embodiment, the operation center of the industrial park microgrid first uses the data modeling module of this invention to obtain complete operational data of the microgrid, including detailed model parameters, line parameters, and real-time topology of the 5MW photovoltaic array, 3MW wind farm, 2MW / 4MWh energy storage device, and large induction motors on multiple production lines within the park. Based on this data, the system automatically constructs a nonlinear differential equation model containing 158 state variables, and performs linearization processing at the current operating point, i.e., a total load of 4.5MW and photovoltaic output of 3.2MW, generating a system state-space model.

[0111] Subsequently, based on this state-space model, the system's sensitivity calculation module established a state trajectory sensitivity equation and solved it for preset disturbance scenarios such as three-phase short-circuit faults on the bus. The calculation results show that the transient sensitivity index of the inverter voltage control loop parameters of the energy storage device is as high as 0.89, the index of the phase-locked loop parameters of the wind power plant converter is 0.72, while the parameter indices of non-critical static loads such as office buildings in the industrial park are generally lower than 0.1.

[0112] The component screening module, with a sensitivity threshold of 0.5, initially identified energy storage inverters and wind power converters as candidate key components. Further analysis of the system state matrix revealed significant dynamic coupling between the state variables of these two components. Therefore, the energy storage device and the wind farm were ultimately determined to constitute the set of key transient impact components affecting this disturbance scenario.

[0113] Next, the model order reduction module, targeting a three-phase short-circuit fault scenario occurring on a critical production line power supply feeder, retained detailed models of the key transient impact components, namely energy storage devices and wind farms. Simultaneously, it performed dynamic equivalent simplification on non-critical components such as office building loads located at greater electrical distances and scattered rooftop photovoltaic systems, constructing a reduced-order dynamic equivalent model for disturbance analysis. The system order of this model was significantly reduced from 158 to 45.

[0114] The time-domain simulation module used this reduced-order model for simulation. The initial simulation conditions were determined by the system's power flow calculations. For example... Figure 4 In the simulation, a three-phase short circuit occurred on the line at t=0.1s, and the fault was cleared at t=0.25s. The simulation process recorded the dynamic response data of the system frequency and the voltage of the critical load bus. As shown in the figure, the trajectories of the two curves are highly consistent during the disturbance occurrence (t=0.1s), fault clearing (t=0.25s), and the subsequent dynamic recovery process. This demonstrates that the model reduction method proposed in this invention can reproduce the key dynamic characteristics of the system without distortion, thus verifying that this invention can improve computational efficiency while ensuring the accuracy of transient stability analysis results. The analysis report generation module judges the system stability based on the convergence of the simulation data and gradually increases the fault duration through iterative simulation, finally calculating the transient stability margin of the system, i.e., the critical clearing time (CCT), to be 162ms.

[0115] By adjusting model parameters using actual operational data and constructing a system state-space model using the updated data, the analysis is ensured to start with a high-fidelity mathematical model that accurately reflects the dynamic characteristics of a specific microgrid. By identifying key components that significantly influence the transient processes of the target microgrid and simplifying only the non-critical parts, the model's complexity is reduced while preserving crucial dynamic information, thus improving the computational efficiency of time-domain simulation and enabling rapid analysis across large scales and scenarios. Quantitative safety assessments are conducted by calculating transient stability margins, and correlation analysis between stability margins and key transient impact components reveals the root causes affecting system stability and generates specific and feasible system optimization suggestions. This provides clear decision support for operators and gives the analysis results practical engineering value.

[0116] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0117] The following are device embodiments of the present invention. For details not described in detail, please refer to the corresponding method embodiments described above.

[0118] Figure 5 A schematic diagram of the microgrid transient stability analysis device provided in an embodiment of the present invention is shown. For ease of explanation, only the parts related to the embodiment of the present invention are shown, and are described in detail below: like Figure 5 As shown, the microgrid transient stability analysis device 5 includes: The update module 51 is used to update the model parameters of each component in the target microgrid based on the operating data of the target microgrid, so as to obtain the updated model parameters of each component. Module 52 is used to construct the system state-space model of the target microgrid based on the operating data of the target microgrid and the updated model parameters of each component. The determination module 53 is used to determine the transient sensitivity index of each component within the transient time window based on the preset disturbance amount and the system state space model. Analysis module 54 is used to perform transient stability analysis on the target microgrid based on the transient sensitivity index of each component, the operating data of the target microgrid, and the system state-space model, and to obtain the transient stability analysis results of the target microgrid.

[0119] In one possible implementation, the update module 51 is specifically used to: extract the time series data of the target microgrid after disturbance from the operating data of the target microgrid, and determine the time series data as the actual system response; perform time-domain simulation of the target microgrid using multiple sets of candidate parameters of all components to obtain multiple simulated system responses of the target microgrid; calculate the error between each simulated system response and the actual system response of the target microgrid according to a preset objective function; and determine the set of candidate parameters corresponding to the simulated system response with the smallest error as the updated model parameters of each component.

[0120] In one possible implementation, the construction module 52 is specifically used to: construct the power flow equations of the target microgrid and the differential equations of each component based on the operating data of the target microgrid and the updated model parameters of each component; determine the precise parameterized model of the target microgrid based on the differential equations of each component; and determine the system state-space model of the target microgrid based on the precise parameterized model and the power flow equations.

[0121] In one possible implementation, the construction module 52 is further configured to: solve the power flow equations using the precise parameterized model to obtain the precise equilibrium point of the target microgrid; and at the precise equilibrium point, linearize the precise parameterized model to obtain the system state-space model of the target microgrid.

[0122] In one possible implementation, module 53 is specifically used for: applying a preset perturbation amount to the updated model parameters of each component to obtain a perturbed system state-space model; calculating partial derivatives on both sides of each state-space equation in the perturbed system state-space model to obtain the state trajectory sensitivity equation of each component; solving the state trajectory sensitivity equation of each component using numerical integration and the system state-space model to obtain the sensitivity numerical trajectory of each component; for each component, extracting the sensitivity of the component at each moment within the transient time window from the sensitivity numerical trajectory of the component, and performing a weighted integral on the sensitivity at each moment within the transient time window to obtain the transient sensitivity index of the component.

[0123] In one possible implementation, the analysis module 54 is specifically used for: selecting multiple key components from all components of the target microgrid based on the transient sensitivity index of each component and the system state-space model; reducing the order of the simulation model of the target microgrid according to the target disturbance scenario and multiple key components to obtain the dynamic equivalent model of the disturbance analysis of the target microgrid; controlling the dynamic equivalent model of the disturbance analysis of the target microgrid to perform simulation according to the operating data of the target microgrid to obtain the time-domain simulation data of the target microgrid; when the dynamic response curve in the time-domain simulation data converges, controlling the equivalent model of the disturbance analysis to perform iterative simulation to obtain the transient stability margin of the target microgrid; performing correlation analysis on the transient stability margin of the target microgrid and multiple key components to obtain the correlation analysis results; obtaining system optimization suggestions for the target microgrid based on the correlation analysis results; and determining the transient stability margin and system optimization suggestions of the target microgrid as the transient stability analysis results of the target microgrid.

[0124] In one possible implementation, the analysis module 54 is further configured to: identify the component corresponding to the transient sensitivity index that exceeds the sensitivity threshold as a candidate key component; identify the value corresponding to the off-diagonal element in the system state matrix of the system state space model as the coupling strength of the two components corresponding to the off-diagonal element; when the coupling strength between the first component and the second component exceeds a preset coupling threshold, and the first component is a candidate key component while the second component is not a candidate key component, identify the second component as a coupling key component; and identify all candidate key components and coupling key components as key components.

[0125] In one possible implementation, the analysis module 54 is further configured to: identify regions where the electrical distance from each disturbance point in the target disturbance scenario exceeds a preset distance as reduced-order regions; and perform equivalent simplification on all components that are not critical components within the reduced-order regions to obtain a dynamic equivalent model for disturbance analysis of the target microgrid.

[0126] Figure 6This is a schematic diagram of an electronic device provided in an embodiment of the present invention. For example... Figure 6 As shown, the electronic device 6 of this embodiment includes a processor 60 and a memory 61. The memory 61 stores a computer program 62. When the processor 60 executes the computer program 62, it implements the steps in the various method embodiments described above. Alternatively, when the processor 60 executes the computer program 62, it implements the functions of each module / unit in the various device embodiments described above.

[0127] For example, computer program 62 may be divided into one or more modules / units, which are stored in memory 61 and executed by processor 60 to complete the present invention. The one or more modules / units may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of computer program 62 in electronic device 6.

[0128] Electronic device 6 may include, but is not limited to, processor 60 and memory 61. Those skilled in the art will understand that... Figure 6 This is merely an example of electronic device 6 and does not constitute a limitation on electronic device 6. It may include more or fewer components than shown, or combine certain components, or different components. For example, electronic device 6 may also include input / output devices, network access devices, buses, etc.

[0129] For the sake of simplicity and clarity, only the above-described functional modules / units are used as examples. In practical applications, the functions described above can be assigned to different functional modules / units as needed. These modules / units can be implemented in hardware, software, or a combination of both.

[0130] In the above embodiments, the descriptions of each embodiment have their own emphasis. Parts not detailed or described in a particular embodiment can be referred to in the relevant descriptions of other embodiments. Unless otherwise specified or in conflict with logic, the terminology and / or descriptions between different embodiments are consistent and can be referenced interchangeably. Technical features in different embodiments can be combined to form new embodiments based on their inherent logical relationships.

[0131] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A method for transient stability analysis of microgrids, characterized in that, include: The model parameters of each component in the target microgrid are updated based on the operating data of the target microgrid to obtain the updated model parameters of each component. Based on the operating data of the target microgrid and the updated model parameters of each component, a system state-space model of the target microgrid is constructed. Within the transient time window, based on the preset disturbance amount and the system state space model, the transient sensitivity index of each component is determined; Based on the transient sensitivity index of each component, the operating data of the target microgrid, and the system state-space model, a transient stability analysis is performed on the target microgrid to obtain the transient stability analysis results.

2. The microgrid transient stability analysis method according to claim 1, characterized in that, The model parameters of each component in the target microgrid are updated based on the operational data of the target microgrid to obtain the updated model parameters of each component, including: From the operational data of the target microgrid, extract the time series data of the target microgrid after disturbance, and determine the time series data as the actual system response; By utilizing multiple sets of candidate parameters for all components, time-domain simulation of the target microgrid is performed to obtain multiple simulated system responses of the target microgrid; Based on a preset objective function, the error between each simulated system response and the actual system response of the target microgrid is calculated; The set of candidate parameters corresponding to the simulated system response with the smallest error is used to determine the updated model parameters for each component.

3. The microgrid transient stability analysis method according to claim 1, characterized in that, Within the transient time window, based on a preset disturbance amount and the system state-space model, the transient sensitivity index of each component is determined, including: Apply a preset perturbation amount to the updated model parameters of each component to obtain the perturbed system state-space model; By taking the partial derivatives on both sides of each state-space equation in the state-space model of the system after the disturbance, the state trajectory sensitivity equation of each component is obtained. Using numerical integration and the system state-space model, the state trajectory sensitivity equation of each component is solved to obtain the sensitivity numerical trajectory of each component. For each component, the sensitivity of the component at each moment within the transient time window is extracted from the sensitivity value trajectory of the component, and the sensitivity at each moment within the transient time window is weighted and integrated to obtain the transient sensitivity index of the component.

4. The microgrid transient stability analysis method according to claim 1, characterized in that, The step of constructing a system state-space model of the target microgrid based on its operational data and the updated model parameters of each component includes: Based on the operating data of the target microgrid and the updated model parameters of each component, the power flow equations of the target microgrid and the differential equations of each component are constructed. Based on the differential equations of each component, a precise parameterized model of the target microgrid is determined; Based on the precise parameterized model and the power flow equation, the system state-space model of the target microgrid is determined.

5. The microgrid transient stability analysis method according to claim 4, characterized in that, The step of determining the system state-space model of the target microgrid based on the precise parameterized model and the power flow equation includes: Using the precise parameterized model, the power flow equations are solved to obtain the precise equilibrium point of the target microgrid; At the precise equilibrium point, the precise parameterized model is linearized to obtain the system state-space model of the target microgrid.

6. The microgrid transient stability analysis method according to claim 1, characterized in that, The transient stability analysis of the target microgrid is performed based on the transient sensitivity index of each component, the operating data of the target microgrid, and the system state-space model, to obtain the transient stability analysis results of the target microgrid, including: Based on the transient sensitivity index of each component and the system state-space model, several key components are selected from all components of the target microgrid. Based on the target disturbance scenario and multiple key components, the simulation model of the target microgrid is reduced in order to obtain a dynamic equivalent model for disturbance analysis of the target microgrid. The dynamic equivalent model for disturbance analysis of the target microgrid is controlled, and simulation is performed according to the operating data of the target microgrid to obtain the time-domain simulation data of the target microgrid; When the dynamic response curve in the time-domain simulation data converges, the disturbance analysis equivalent model is controlled to perform iterative simulation to obtain the transient stability margin of the target microgrid. Correlation analysis was performed on the transient stability margin of the target microgrid and several key components to obtain the correlation analysis results; Based on the correlation analysis results, system optimization suggestions for the target microgrid are obtained; The transient stability margin and system optimization suggestions of the target microgrid are determined as the transient stability analysis results of the target microgrid.

7. The microgrid transient stability analysis method according to claim 6, characterized in that, Based on the transient sensitivity index of each component and the system state-space model, several key components are selected from all components of the target microgrid, including: Components corresponding to transient sensitivity indicators that exceed the sensitivity threshold are identified as candidate key components; The values ​​corresponding to the off-diagonal elements in the system state matrix of the system state space model are determined as the coupling strength between the two components corresponding to the off-diagonal elements. When the coupling strength between the first component and the second component exceeds a preset coupling threshold, and the first component is a candidate key component while the second component is not a candidate key component, the second component is identified as the coupling key component. All candidate key components and coupling key components are identified as key components.

8. The microgrid transient stability analysis method according to claim 6, characterized in that, The step of reducing the order of the simulation model of the target microgrid based on the target disturbance scenario and multiple key components to obtain a dynamic equivalent model for disturbance analysis of the target microgrid includes: The region whose electrical distance from each disturbance point in the target disturbance scenario exceeds a preset distance is defined as the order reduction region; For all components that are not critical components in the reduced-order region, equivalent simplification is performed to obtain the dynamic equivalent model for disturbance analysis of the target microgrid.

9. A microgrid transient stability analysis device, characterized in that, include: The update module is used to update the model parameters of each component in the target microgrid based on the operating data of the target microgrid, so as to obtain the updated model parameters of each component. The construction module is used to construct the system state-space model of the target microgrid based on the operating data of the target microgrid and the updated model parameters of each component; The determination module is used to determine the transient sensitivity index of each component within a transient time window, based on a preset disturbance amount and the system state space model. The analysis module is used to perform transient stability analysis on the target microgrid based on the transient sensitivity index of each component, the operating data of the target microgrid, and the system state-space model, and to obtain the transient stability analysis results of the target microgrid.

10. An electronic device, characterized in that, It includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the method as described in any one of claims 1 to 8.

Images