Voltage source feature feasible region quantization method and system for a data mapping device

Abstract

This invention discloses a method and system for quantifying the feasible region of voltage source characteristics in a data mapping device, belonging to the field of power system operation and control technology. The method includes: collecting electrical topology and control loop parameters, merging them to obtain a set of operating parameters; dividing these parameters into slow dynamic quantities of the control loop and fast dynamic quantities of network electromagnetic transients, establishing dynamic sub-equations for each and performing algebraic reduction to construct a decoupled dynamic model; deriving the analytical expression of transient currents, merging constraints and eliminating redundancy to generate a set of voltage source characteristic constraints; obtaining a set of feasible region vertices through boundary vertex search and filtering, forming the feasible region boundary through adjacency reconstruction and convex hull closure; finally, quantifying the boundary volume and operating margin, mapping it to the output of the full-condition voltage source characteristic operating boundary. This invention employs a slow-fast variable hierarchical decoupling mechanism, combined with convex hull reconstruction and quantization analysis, to achieve a quantitative characterization of the feasible region of voltage source characteristics.

Description

Technical Field

[0001] This invention relates to the field of power system operation control technology, and in particular to a method and system for quantizing the feasible domain of voltage source characteristics of a data mapping device. Background Technology

[0002] With the rapid development of new energy power generation technologies, voltage source power electronic devices, represented by grid-connected energy storage devices, are widely used in power systems. These devices, by simulating the external characteristics of synchronous generators, can actively construct and support grid voltage and frequency, which is crucial for ensuring the stable operation of power supply circuits and systems. The dynamic performance of grid-connected devices is highly dependent on their internal control loop parameters and the electrical topology parameters of the external power grid. The coordinated operation of these parameters directly determines whether they can maintain stable voltage source characteristics under various disturbances. The data mapping device is a supporting data processing unit for grid-connected energy storage devices or voltage source power electronic devices. It is used to collect grid electrical topology parameters and device control loop parameters, perform parameter merging, per-unit processing, and dynamic characteristic layering, providing standardized input for decoupled dynamic modeling. It is the core data interface module for realizing the quantification of the feasible region of voltage source characteristics.

[0003] In related technologies, Chinese invention patent CN114977297B discloses a method, apparatus, and terminal equipment for constructing a feasible region of virtual inertia and damping parameters, including: constructing an equivalent model of a microgrid and determining the transfer function of the equivalent model; establishing multiple constraints based on the transfer function and load disturbance, wherein each constraint includes virtual inertia and damping coefficient; and analyzing the multiple constraints based on multiple preset thresholds corresponding to the multiple constraints to obtain a first feasible region of virtual inertia and damping coefficient.

[0004] However, the aforementioned existing technical solutions have the following technical shortcomings: they only focus on two types of control loop parameters—virtual inertia and damping coefficient—and do not include electrical topology parameters such as bus voltage and line impedance. This prevents the coordinated evaluation of electrical topology parameters and control loop parameters, limiting the ability to understand and regulate the overall stability of the power system. They only establish constraints based on a single load disturbance condition, failing to cover all operating scenarios such as startup, grid connection, and fault ride-through. The resulting feasible region cannot adapt to the operational needs of multiple scenarios, and the characterization of the feasible region for voltage source characteristics is not comprehensive or accurate enough. They do not employ a hierarchical decoupling mechanism for slow and fast variables, resulting in high complexity and slow computation speed for system models containing dynamic characteristics across multiple time scales, making it difficult to meet the needs of real-time operation control of the power system. Finally, they fail to transform complex stability analysis results into intuitive quantitative indicators and queryable operational boundary tools, thus failing to provide clear operational guidance for grid dispatchers and limiting their practical value. Summary of the Invention

[0005] To address the aforementioned issues, this invention provides a method and system for quantizing the feasible region of voltage source features in a data mapping device. It employs a slow-fast variable hierarchical decoupling mechanism, combined with convex hull reconstruction and quantization analysis, to achieve a quantitative characterization of the feasible region of voltage source features.

[0006] The above objectives can be achieved through the following approach:

[0007] A method and system for quantifying the feasible region of voltage source characteristics in a data mapping device includes: acquiring electrical topology and control loop parameters, merging them to obtain a set of operating parameters; dividing these parameters into slow dynamic quantities of the control loop and fast dynamic quantities of network electromagnetic transients, establishing dynamic sub-equations for each and performing algebraic reduction to construct a decoupled dynamic model; deriving analytical expressions for transient currents, merging constraints and eliminating redundancy to generate a set of voltage source characteristic constraints; obtaining a set of feasible region vertices through boundary vertex search and filtering, forming a feasible region boundary through adjacency reconstruction and convex hull closure; and finally quantifying the boundary by volume measurement and operating margin, mapping it to the output of the full-condition voltage source characteristic operating boundary.

[0008] Optionally, the acquisition of electrical topology parameters and control loop parameters, and the merging and integration of the acquisition results according to physical quantity categories to obtain the operating parameter set, includes: reading bus voltage phasors, branch current phasors, and switch position quantities, and arranging them according to topology node numbers to obtain electrical topology parameters; extracting droop coefficients, virtual inertia coefficients, and voltage outer loop bandwidth quantities, and labeling them according to control levels to obtain control loop parameters; and merging and integrating the electrical topology parameters and the control loop parameters according to physical quantity categories and performing unit standardization processing to obtain the operating parameter set.

[0009] Optionally, the step of hierarchically dividing the set of operating parameters to obtain the slow dynamic quantity of the control loop and the fast dynamic quantity of the network electromagnetic transient includes: identifying the characteristic frequency bands of each physical quantity in the set of operating parameters to obtain a parameter subsequence with frequency band identifiers; decomposing the parameter subsequence into low-frequency components and high-frequency components according to the frequency band identifiers to obtain a hierarchical parameter sequence; aggregating the low-frequency components in the hierarchical parameter sequence according to the control loop response window to obtain the slow dynamic quantity of the control loop; and truncating the high-frequency components in the hierarchical parameter sequence according to the electromagnetic transient time window to obtain the fast dynamic quantity of the network electromagnetic transient.

[0010] Optionally, the construction of the decoupled dynamic model includes: constructing a control loop state equation containing droop terms and virtual inertia terms based on the slow dynamics of the control loop, to obtain a control loop dynamic sub-equation; constructing a transient network equation containing the relationship between inductor-capacitor branches and node admittances based on the network electromagnetic transient fast dynamics, to obtain a network dynamic sub-equation; and performing slow-fast variable algebraization order reduction and boundary matching processing on the control loop dynamic sub-equation and the network dynamic sub-equation according to the time scale difference to obtain the decoupled dynamic model.

[0011] Optionally, the step of loading excitations for all types of disturbance conditions based on the decoupled dynamic model, solving the transient response trajectories for each condition, and extracting explicit expressions containing operating parameters to obtain the analytical expression of transient current includes: setting a set of disturbance conditions including startup, grid connection, load disturbance, and fault ride-through; loading the set of disturbance conditions as excitation inputs item by item into the decoupled dynamic model to obtain the transient response trajectories for each condition; performing time-domain integral transformation and algebraic elimination on the transient response trajectories for each condition in sequence, and extracting explicit expressions regarding operating parameters to obtain the analytical expression of transient current.

[0012] Optionally, generating the voltage source feature constraint set includes: obtaining the voltage source feature maintenance conditions; analyzing the voltage source feature maintenance conditions according to three dimensions: voltage amplitude, frequency, and phase angle, to obtain voltage source feature indices; reconstructing the voltage source feature indices using inequalities in conjunction with the transient current analytical expression, to obtain a family of constraint inequalities regarding operating parameters; and combining the family of constraint inequalities according to operating condition identifiers and removing redundant constraints to generate the voltage source feature constraint set.

[0013] Optionally, generating the parameter feasible domain vertex set includes: solving each constraint surface in the voltage source feature constraint set to obtain a candidate vertex set; substituting each vertex in the candidate vertex set back into the voltage source feature constraint set for feasibility verification, removing vertices that violate any constraint to obtain a valid vertex set; and clustering the valid vertex set according to the running parameter dimension and merging duplicate points to generate the parameter feasible domain vertex set.

[0014] Optionally, forming the parameter feasible region boundary includes: calculating the vertices in the vertex set of the parameter feasible region to obtain vertex adjacency relationships; connecting adjacent vertices sequentially according to the nearest neighbor principle to obtain a feasible region boundary polyline; performing convex hull closure processing on the feasible region boundary polyline, and identifying and marking the internal closed-loop boundaries through constraint surface analysis to form the parameter feasible region boundary.

[0015] Optionally, the step of performing volume measurement and operational margin quantization calculations on the feasible region boundary of the parameters to obtain a feasible region quantization result, and then mapping the feasible region quantization result to the characteristic operational boundary of the voltage source under all operating conditions, includes: performing volume measurement on the feasible region boundary of the parameters using random sampling integration to obtain a feasible region capacity index; calculating the target normal distance from the current operating point of the grid-type energy storage device to the feasible region boundary of the parameters to obtain a feasible region margin index; fusing the feasible region capacity index and the feasible region margin index according to operating condition weights to obtain a feasible region quantization result; and organizing the feasible region quantization result into an operational boundary lookup table according to the operating parameter dimension, which is then output as the characteristic operational boundary of the voltage source under all operating conditions.

[0016] Based on the same inventive concept, this invention also provides a voltage source characteristic feasible region quantization system for a data mapping device. The system includes: a parameter acquisition and preprocessing module for acquiring electrical topology parameters and control loop parameters, and merging and integrating the acquired results according to physical quantity categories to obtain a set of operating parameters; a dynamic characteristic hierarchical module for hierarchically dividing the set of operating parameters to obtain slow dynamic quantities of the control loop and fast dynamic quantities of the network electromagnetic transients; a decoupling dynamics modeling module for establishing dynamic sub-equations based on the slow dynamic quantities of the control loop and the fast dynamic quantities of the network electromagnetic transients, performing slow-fast variable algebraic reduction on the dynamic sub-equations to construct a decoupling dynamics model; and a transient current analysis module for loading excitations of all types of disturbance conditions based on the decoupling dynamics model and solving for the transient response under each condition. The transient current is analyzed by extracting an explicit expression containing operating parameters from the trajectory through integral transformation and algebraic elimination. A feature constraint generation module maps the transient current to a constraint, merges constraints, and removes redundancy to generate a voltage source feature constraint set. A feasible region vertex search module searches for boundary vertices and performs feasibility screening on the voltage source feature constraint set to generate a parameter feasible region vertex set. A feasible region boundary reconstruction module reconstructs the parameter feasible region vertex set based on vertex adjacency relationships and performs convex hull closure to form the parameter feasible region boundary. A quantization evaluation and output module performs quantization calculations on the parameter feasible region boundary for volume measurement and operating margin, obtains the feasible region quantization result, and maps the feasible region quantization result to the full-condition voltage source feature operating boundary before outputting it.

[0017] Compared with the prior art, the present invention has the following advantages:

[0018] This invention improves the efficiency and speed of analyzing the feasible region of voltage source characteristics by decoupling dynamic modeling and geometric reconstruction.

[0019] This invention systematically combines multidimensional parameters with multi-condition constraints to comprehensively and accurately characterize the safe operating boundary of the equipment.

[0020] This invention transforms complex stability analysis results into intuitive quantitative indicators and operational boundary lookup tables, providing clear operational guidance for power grid dispatching.

[0021] This invention enables the coordinated evaluation of electrical topology parameters and control loop parameters, thereby enhancing the understanding and control capabilities of the overall stability of the power system.

[0022] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures pointed out in the description, claims and drawings. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 This is a flowchart illustrating a feasible domain quantization method for voltage source features of a data mapping device according to an embodiment of the present invention.

[0025] Figure 2 This is the response surface plot of the decoupled dynamic model according to an embodiment of the present invention.

[0026] Figure 3 This is a diagram showing the results of feasible domain vertex search and filtering in an embodiment of the present invention.

[0027] Figure 4 This is a multi-dimensional comparison chart of the feasible domain quantification results of an embodiment of the present invention.

[0028] Figure 5 This is a schematic diagram of the structure of a voltage source feature feasible domain quantization system of a data mapping device according to an embodiment of the present invention. Detailed Implementation

[0029] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0030] Reference Figure 1One embodiment of the present invention proposes a method for quantizing the feasible region of voltage source features in a data mapping device. It adopts a slow-fast variable hierarchical decoupling mechanism and combines convex hull reconstruction and quantization analysis to achieve quantization characterization of the feasible region of voltage source features.

[0031] The method described in this embodiment specifically includes:

[0032] S1. Collect electrical topology parameters and control loop parameters, and merge and integrate the collected results according to the category of physical quantities to obtain a set of operating parameters;

[0033] Optionally, the acquisition of electrical topology parameters and control loop parameters, and the merging and integration of the acquisition results according to physical quantity categories, yield a set of operating parameters including:

[0034] Read the bus voltage phasor, branch current phasor, and switch position, and arrange them according to the topology node number to obtain the electrical topology parameters;

[0035] Extract the droop coefficient, virtual inertia coefficient, and voltage outer loop bandwidth, and label them according to the control level to obtain the control loop parameters;

[0036] The electrical topology parameters and the control loop parameters are merged and integrated according to physical quantity categories and then standardized to obtain a set of operating parameters.

[0037] Specifically, data is collected from real-time monitoring systems of the power grid, such as SCADA or WAMS, via data interfaces. The system periodically reads and records bus voltage phasors, branch current phasors, and switch positions at specific time points. For example, in a 10kV distribution network, the device collects the voltage amplitude and phase angle at each node on all feeders, the current amplitude and phase angle flowing through each line, and the on / off status of all circuit breakers and disconnectors. This data is structured according to a preset power grid topology node numbering to form an electrical topology parameter matrix, ensuring consistency between the data and the physical topology.

[0038] The system acquires key control settings within the voltage source controller, which directly determine the device's dynamic response characteristics. The system accesses the controller backend of voltage source devices such as grid-connected energy storage devices via a communication bus, or extracts parameters directly from the device configuration files. Key extracted control loop parameters include the droop coefficient, which defines the static relationship between active power and frequency, typically between 2% and 5%; the virtual inertia coefficient, which determines the device's instantaneous response to frequency changes and is used to simulate the inertia of a synchronous generator; and the voltage outer loop bandwidth, which reflects the response speed of the voltage control loop and determines the dynamic performance of voltage recovery. The extracted parameters are bound to the corresponding device ID and labeled according to control level, such as primary control, secondary control, etc., forming a control loop parameter list.

[0039] The system integrates parameters from different sources and using different physical units into a dataset. It merges the electrical topology parameters and control loop parameters obtained in the first two steps. Then, it performs unit normalization, which requires pre-setting system baseline values, such as baseline capacity and baseline voltage. All physical quantities are converted into ratios relative to their baseline values. The normalization calculation follows the formula:

[0040] ,

[0041] in Per unit value, The actual measured value or constant value of a physical quantity. This is the reference value corresponding to the physical quantity. For example, an actual voltage of 10.5kV has a per-unit value of 1.05 under a 10kV reference. Through this process, all parameters are unified into dimensionless per-unit values.

[0042] The data mapping device outputs a structured set of operating parameters. This set is a multi-dimensional vector that has been merged and normalized, integrating electrical topology parameters characterizing the power grid's operating state with control loop parameters determining the dynamic behavior of equipment. All elements in the set are dimensionless values ​​and are clearly indexed and labeled according to the physical topology nodes and control equipment levels, providing direct and standardized data input for the subsequent construction of a decoupled dynamic model.

[0043] For example, the acquisition of parameters for a grid-connected energy storage device in a 10kV distribution network is used as an example for reproduction. The data mapping device is a supporting data acquisition and per-unit unit for grid-connected energy storage. It is responsible for merging electrical topology and control loop parameters into a standardized set of operating parameters and serves as the data entry point and execution carrier for the feasible domain quantification method of this invention. First, the system reads the bus voltage phasor, branch current phasor, and switch position from the power grid dispatch system through the data interface, and arranges the collected data in a structured manner according to the grid topology node number to form electrical topology parameters. Assume that the bus voltage phasor read at node number 01 is 10.5kV, the branch current phasor is 100A, and the switch position is 1. Subsequently, the system accesses the controller backend to extract key control loop parameters, obtaining the device's droop coefficient as 0.03, virtual inertia coefficient as 5s, and voltage outer loop bandwidth as 50Hz, and annotates these parameters according to the primary control level. Finally, the system merges and integrates the acquired electrical topology parameters and control loop parameters according to physical quantity categories and performs per-unit standardization. Set the reference voltage to 10kV, calculate the per-unit value of the bus voltage. Through the above processing, the device outputs a dimensionless set of operating parameters containing topological state and control dynamics, providing a standardized data index for subsequent decoupled dynamics modeling.

[0044] S2. The set of operating parameters is divided into hierarchical components to obtain the slow dynamics of the control loop and the fast dynamics of the network electromagnetic transient.

[0045] Optionally, the step of hierarchically dividing the set of operating parameters to obtain the slow dynamics of the control loop and the fast dynamics of the network electromagnetic transients includes:

[0046] Characteristic frequency bands are identified for each physical quantity in the set of operating parameters to obtain parameter subsequences with frequency band identifiers;

[0047] Based on the frequency band identifier, the parameter subsequence is decomposed into low-frequency components and high-frequency components to obtain a hierarchical parameter sequence;

[0048] The low-frequency components in the hierarchical parameter sequence are aggregated over time according to the control loop response window to obtain the slow dynamics of the control loop.

[0049] The high-frequency components in the hierarchical parameter sequence are extracted according to the electromagnetic transient time window to obtain the network electromagnetic transient fast dynamic quantity.

[0050] Specifically, the data mapping device performs characteristic frequency band identification on the time-series data of each parameter in the operating parameter set, such as the fluctuation sequence of bus voltage or virtual inertia coefficient. This identification process typically employs spectral analysis algorithms such as Fast Fourier Transform (FFT) or wavelet transform to convert the time-domain signal into a frequency-domain signal. The system presets a cutoff frequency, which is the boundary distinguishing between electromechanical dynamics and electromagnetic transients; in engineering, it is usually set between 50 Hz and 150 Hz. By analyzing the spectral energy distribution of each parameter sequence, if its main energy is concentrated below the cutoff frequency, it is identified as a low-frequency component; conversely, if the energy is mainly distributed above the cutoff frequency, it is identified as a high-frequency component. The output of this step is a parameter subsequence with frequency band identification.

[0051] The system extracts and aggregates slow dynamic quantities that represent the steady-state behavior of the control system from the hierarchical parameter sequence. Based on the frequency band identification from the previous step, the system decomposes the parameter subsequence into low-frequency and high-frequency components, forming a hierarchical parameter sequence. For the low-frequency components, the system performs time aggregation. Time aggregation calculates the average or integral value within a preset control loop response window to smooth out minor fluctuations and extract trend changes. The time length of the control loop response window is specified. The response window is typically determined based on the controller's setpoint; for example, for droop control, the response window may be between 0.5 and 5 seconds. The aggregate calculation can be expressed as the formula:

[0052] ,

[0053] In the time interval Integrate on top. Among them... It is the slow dynamic quantity of the control loop obtained after aggregation. It is the low-frequency component in the original parameter sequence. This is the duration of the control loop response window. The output of this process is the control loop slow dynamics, which represents the controller's slow adjustment behavior, such as power distribution and frequency adjustment.

[0054] The system accurately captures the electromagnetic transient behavior of the power grid during disturbances, providing high-frequency data for transient stability analysis. It extracts the portion of the hierarchical parameter sequence identified as high-frequency components. This extraction is triggered by the system disturbance monitoring module, for example, when a fault current or voltage surge is detected. The system uses the time of disturbance occurrence as the basis for its analysis. Starting from a preset electromagnetic transient time window Within this timeframe, the waveform data of high-frequency components are completely preserved. The length of the electromagnetic transient time window... Typically short, ranging from 20 to 200 milliseconds, sufficient to cover the initial few power frequency cycles after a fault occurs. The interception operation can be described as follows:

[0055] ,

[0056] when In the interval Inside. Among them It is the network electromagnetic transient fast dynamic quantity obtained after interception. These are the high-frequency components in the original parameter sequence. This is the moment the disturbance is triggered. The output of this step is the network electromagnetic transient fast dynamic quantity, which accurately records the rapid voltage and current oscillations of the lines and equipment under disturbance.

[0057] The hierarchical partitioning process outputs two independent sets of dynamic quantities: slow dynamic quantities of the control loop and fast dynamic quantities of the network electromagnetic transients. The former reflects the quasi-steady-state and electromechanical dynamic characteristics of the system, while the latter accurately characterizes the rapid electromagnetic oscillation process in the early stages of a disturbance. These two sets will serve as independent inputs for the subsequent construction of a decoupled dynamic model, achieving effective separation of complex dynamic processes along the time scale.

[0058] For example, the data mapping device first uses Fast Fourier Transform to identify the characteristic frequency bands of each physical quantity in the aforementioned set of operating parameters, obtaining parameter sub-sequences with frequency band identifiers. The device sets the cutoff frequency to 100Hz and identifies parameter sequences above this value as high-frequency components and those below as low-frequency components, thereby obtaining a hierarchical parameter sequence. For low-frequency components, the device operates within a preset control loop response window. Time aggregation is performed internally using the formula. The slow dynamics of the control loop are calculated. If the total value obtained by integrating a low-frequency voltage fluctuation sequence within a window is 0.4 per-unit seconds, then the aggregated slow dynamics of the control loop are... For high-frequency components, the device detects the fault triggering time. Then, cut off Waveform data within the electromagnetic transient time window, using the formula ,exist The real-time electromagnetic transient response of the high-frequency component is completely extracted within the interval. This step ultimately outputs two independent sets of dynamic quantities, realizing the effective stratification of the system's electromechanical and electromagnetic characteristics.

[0059] S3. Based on the slow dynamics of the control loop and the fast electromagnetic transients of the network, establish dynamic sub-equations respectively, perform slow and fast variable algebraization to reduce the order of the dynamic sub-equations, and construct a decoupled dynamic model;

[0060] Optionally, the construction of the decoupled dynamic model includes:

[0061] Based on the slow dynamics of the control loop, a control loop state equation containing a droop term and a virtual inertia term is constructed, and the control loop dynamic sub-equation is obtained.

[0062] Based on the network electromagnetic transient fast dynamics, a transient network equation including the relationship between inductor-capacitor branches and node admittance is constructed, and the network dynamics sub-equation is obtained;

[0063] The control loop dynamics sub-equations and the network dynamics sub-equations are reduced in order and subjected to boundary matching by slow-fast variable algebraization based on the time scale difference, resulting in a decoupled dynamics model.

[0064] Specifically, a mathematical equation capable of accurately describing the dynamic response of the controller is constructed. The data mapping device utilizes the slow dynamic quantities of the control loop obtained in the previous step to establish the control loop dynamic sub-equations. This equation is a system of ordinary differential equations, the core of which is an equation simulating the generator's power angle oscillation characteristics, integrating droop control and virtual inertia. Its typical form is:

[0065] ,

[0066] in It is the virtual inertia coefficient, used to characterize the system's dynamic support capability for frequency changes. Its value is generally in the range of 2 to 8 seconds and is obtained from the control loop parameters. It is the angular frequency of the system, and the state variable of the model. It is the equivalent mechanical power, determined by the droop control characteristics. This is the output electromagnetic power, serving as the coupling interface with the network sub-equations. These equations describe how the control system adjusts the power to respond to frequency deviations on a second-scale timescale.

[0067] A mathematical model capable of precisely characterizing the electromagnetic transient processes of a power grid is established. Based on the fast dynamic variables of network electromagnetic transients, the system constructs network dynamics sub-equations. This set of equations consists of differential equations describing the dynamics of inductive and capacitive elements and algebraic equations describing network topological constraints, typically expressed in the form of nodal admittance matrices.

[0068] ,

[0069] in, and These are the injected current vector and the node voltage vector of the network node, respectively, extracted from the fast dynamic quantities of the network electromagnetic transients. It is a nodal admittance matrix calculated based on electrical topology parameters, containing information such as line impedance. This set of equations accurately describes the rapid oscillation and propagation behavior of grid voltage and current on a millisecond timescale.

[0070] The two sub-models with different time scales are integrated into a unified and solvable decoupled dynamic model. The system employs slow-fast variable algebraic reduction and boundary matching techniques. The criterion for slow-fast variable algebraic reduction is the difference in time scales: when the time constant of the network's electromagnetic transient fast dynamics... Time constant of slow dynamics in the control loop satisfy:

[0071] ,

[0072] The fast dynamic process can be approximated as an algebraic constraint, and its differential terms can be ignored to achieve order reduction. The specific rule is as follows: when analyzing the slow dynamic process, it is assumed that the fast dynamic process has instantaneously reached a transient steady state, and the differential terms in the network dynamics sub-equations are... and Setting the value to zero reduces the network sub-equations to a system of pure algebraic equations. When analyzing fast dynamic processes, slow dynamic variables are treated as constants that remain unchanged within the short time window. Boundary matching is achieved through coupling variables. Achieving bidirectional interaction: Calculated from the dynamic sub-equations of the control loop As the input power boundary condition of the network dynamics sub-equation, it is substituted into the formula to participate in the network power flow calculation; the grid connection point voltage obtained by solving the network dynamics sub-equation Current Feedback is fed back to the control loop equations to update the reference values ​​for droop control and virtual inertia adjustment, thereby achieving continuous matching of slow and fast dynamics at the boundaries. Through the above processing, the data mapping device outputs a decoupled dynamic model, and the response surface plot of the decoupled dynamic model is shown below. Figure 2As shown, the decoupled dynamics model consists of two parts: the control loop slow dynamic equations and the network fast dynamic equations. The control loop slow dynamic equations employ control loop state equations that include droop and virtual inertia terms. Specifically, the virtual inertia coefficient multiplied by the derivative of the angular frequency with respect to time equals the difference between mechanical power and electromagnetic power. The virtual inertia coefficient is taken from control loop parameters, with a value ranging from two to eight seconds. Electromagnetic power serves as a coupling variable with the network sub-equations. The network fast dynamic equations employ transient network equations that include the relationship between inductor-capacitor branches and node admittance. Specifically, the node injected current vector equals the node admittance matrix multiplied by the node voltage vector. The node admittance matrix is ​​calculated from electrical topology parameters and includes line impedance information. The two equations achieve boundary matching through the coupling variable electromagnetic power: the electromagnetic power calculated by the control loop equations serves as the input power boundary condition for the network equations in power flow calculations. The grid connection point voltage and current obtained from solving the network equations are fed back to the control loop equations to update the reference values ​​for droop control and virtual inertia adjustment, thereby achieving continuous matching between slow and fast dynamics at the boundary, forming a complete decoupled dynamics model.

[0073] For example, the data mapping device constructs a dynamic equation describing the system behavior based on the control loop slow dynamics and network electromagnetic transient fast dynamics obtained in the aforementioned steps. The device first uses the control loop slow dynamics to construct a control loop dynamic sub-equation containing a droop term and a virtual inertia term. Let the virtual inertia coefficient of the energy storage device be... Current mechanical power command Per-unit value, electromagnetic power output The per-unit value is obtained by the system. Solving for the rate of change of frequency Simultaneously, the device constructs transient network equations based on the network's electromagnetic transient fast dynamics, including the relationships between inductor-capacitor branches and node admittances. Let the network node voltage vectors be... Per-unit value, nodal admittance matrix Per-unit value, calculate the injected current vector Per-unit value. Finally, the device treats the control loop dynamics sub-equations and network dynamics sub-equations according to their time scale differences, assuming that the network differential terms are different when analyzing slow dynamic processes. and Setting the value to 0 reduces the network equation to an algebraic equation, and this is achieved through coupling variables. By achieving boundary matching, a decoupled dynamic model is obtained, enabling collaborative modeling of the system's electromechanical dynamics and electromagnetic transient processes.

[0074] S4. Based on the decoupled dynamic model, apply excitation for all types of disturbance conditions, solve the transient response trajectory of each condition, and extract the explicit expression containing the operating parameters through integral transformation and algebraic elimination to obtain the analytical expression of the transient current.

[0075] Optionally, the step of applying excitations for all types of disturbance conditions based on the decoupled dynamic model, solving for the transient response trajectories under different conditions, and extracting explicit expressions containing operating parameters through integral transformation and algebraic elimination to obtain the analytical expression of the transient current includes:

[0076] Set a set of disturbance conditions including startup, grid connection, load disturbance, and fault ride-through;

[0077] The set of disturbance conditions is loaded into the decoupled dynamic model as excitation input item by item to obtain the transient response trajectory of each condition.

[0078] The transient response trajectories under the different operating conditions are sequentially subjected to time-domain integral transformation and algebraic elimination, and explicit expressions for the operating parameters are extracted to obtain the analytical expression of the transient current.

[0079] Specifically, the data mapping device sets a set of disturbance conditions covering typical operating scenarios of voltage source equipment. This set embodies engineering experience and power grid operation specifications, aiming to comprehensively examine the transient response of the equipment under various extreme and normal conditions. This set typically includes: a startup process representing the equipment from standby to operation; a simulation of the dynamic grid connection process at the moment the equipment connects to the grid; load disturbance conditions that test the equipment's ability to cope with normal power fluctuations, such as a step load increase or decrease of 20% of rated capacity; and fault ride-through conditions that assess the equipment's ability to maintain grid connection under severe grid faults, such as simulating a remote three-phase short-circuit fault lasting 100 milliseconds with a voltage drop to 0.2 times the rated value. These conditions together constitute a series of standardized inputs to stimulate the model.

[0080] The device loads each excitation from the set of disturbance conditions into the decoupled dynamic model one by one. Each loading is equivalent to applying a specific initial condition or external input change to the system of differential-algebraic equations described by the model. For example, for the fault ride-through condition, the loading process is equivalent to forcibly setting the voltage at the fault point to a specified drop value in the model. The device then performs numerical or analytical solutions to the model to obtain the dynamic trajectory of the system state variables, especially the transient current, under a specific disturbance, as a function of time, i.e., the transient response trajectory for each condition. Then, the device performs time-domain integral transformation and algebraic elimination on each transient response trajectory for each condition to extract explicit expressions for the operating parameters. The time-domain integral transformation, usually the Laplace transform, is used to convert the differential equations in the time domain into algebraic equations in the frequency domain, which greatly simplifies the solution process. In the frequency domain, the output response... With input stimulus and system transfer function The relationship between them is as follows:

[0081] ,

[0082] Here This represents a set of operating parameters, including the droop coefficient, virtual inertia coefficient, and network impedance. Subsequently, through algebraic elimination and inverse transformation, the device aims to express the peak value of the transient current or its value at a critical moment as operating parameters. The explicit expression. The goal of this process is to obtain a form like The algebraic expression for this. For example, in a simplified second-order system, the peak transient current is expressed as:

[0083] ,

[0084] In this expression, peak current From amplitude Damping ratio Undamped natural oscillation frequency The decision is made based on these parameters, which are calculated from the original operating parameters, such as line impedance and virtual inertia coefficient. , represents the angular frequency of the damped oscillation; Indicates the peak time of the transient current; This represents the initial phase angle. The final function obtained contains the runtime parameters. This is the analytical expression of transient current. It provides the possibility of directly and quickly evaluating the impact of any set of operating parameters on transient current, avoiding repeated numerical simulations of the original complex model.

[0085] For example, the transient current analysis module derives and reproduces the analytical expression of transient current based on the established decoupled dynamics model. First, a set of disturbance conditions including startup, grid connection, load disturbance, and fault ride-through is defined and loaded into the decoupled dynamics model as excitation inputs. Taking the fault ride-through condition as an example, a grid-type energy storage device encounters a remote short-circuit fault, causing a momentary voltage drop at the fault point. This is loaded into the model as an external input change for solution, thus obtaining the transient response trajectory of the transient current changing with time under this condition. Subsequently, time-domain integral transformation and algebraic elimination are sequentially performed on this transient response trajectory. The Laplace transform is introduced as the time-domain integral transform, converting the time-domain differential equation into a frequency-domain algebraic equation. Its complex frequency domain output response satisfies the formula... Assuming that, in the dimensionless operating parameter set obtained by the aforementioned steps, the value of the transfer function frequency domain characteristic matrix calculated from parameters such as line impedance and inertia at a specific frequency point is... The applied voltage drop excitation frequency domain amplitude Substituting into the formula, the frequency domain current response is calculated. Based on this, an inverse transformation and algebraic elimination are performed to extract explicit expressions for the operating parameters. The explicit expression for the peak transient current of this second-order system follows the formula. The system amplitude is then set. Damping ratio Undamped natural oscillation angular frequency initial phase angle Using the formula The damped oscillation angular frequency was calculated. Substitute each transient characteristic parameter into the formula for the peak time of the transient current, and set the peak time. =0.157s. Finally, substituting the above example data completely into the explicit expression for the peak transient current, the result is calculated as follows: Finally, the peak value of the transient current was obtained. Per-unit value. Through the above time-domain integral transformation and algebraic elimination process, the function containing the original operating parameters was successfully extracted, and the accurate reproduction of the analytical expression of transient current under all operating conditions was achieved.

[0086] S5. Perform constraint mapping on the analytical expression of the transient current, merge constraints and remove redundancy to generate a voltage source feature constraint set;

[0087] Optionally, the set of characteristic constraints for generating the voltage source includes:

[0088] Obtain the voltage source characteristic sustaining conditions;

[0089] The voltage source characteristic maintenance conditions are analyzed according to three dimensions: voltage amplitude, frequency, and phase angle, to obtain the voltage source characteristic index.

[0090] By combining the analytical expression of the transient current with the inequality reconstruction of the voltage source characteristic index, a family of constraint inequalities about the operating parameters is obtained;

[0091] The family of constraint inequalities is combined according to the operating condition identifier and redundant constraints are removed to generate a voltage source characteristic constraint set.

[0092] Specifically, the data mapping device acquires the voltage source characteristic maintenance conditions. These conditions are quantitative regulations governing the behavior of grid-connected equipment under various operating conditions, typically derived from power grid guidelines or industry standards. The core requirement is that the equipment must function like an ideal voltage source during transient processes, effectively supporting the grid's voltage and frequency without experiencing loss of synchronization, oscillations, or control instability. A key maintenance condition is that after experiencing severe disturbances such as fault ride-through, the grid connection point voltage and frequency must recover to 90% to 110% of the normal operating range within a specified time, and the current must not exceed 1.2 to 1.5 times the equipment's rated current.

[0093] These macroscopic voltage source characteristic maintenance conditions are analyzed according to three fundamental physical dimensions: voltage amplitude, frequency, and phase angle, resulting in specific voltage source characteristic indicators. For example, for voltage amplitude, the indicator is "the minimum voltage during a fault is not lower than 0.9 pu"; for frequency, the indicator is "the peak transient frequency deviation does not exceed ±0.5 Hz"; and for phase angle, it is "the power angle swing converges within 500 milliseconds, and the maximum power angle difference does not exceed 90 degrees." Furthermore, limitations on transient overcurrents are included, such as "the peak current does not exceed 1.5 pu." These indicators translate system performance requirements into a series of measurable physical quantity boundaries.

[0094] The above voltage source characteristic indices are reconstructed using inequalities to generate a family of constraint inequalities regarding operating parameters. This step is crucial; it utilizes the analytical expressions for transient current and other transient response quantities obtained in the previous step to directly map the constraints of physical quantities to constraints for operating parameters. For example, regarding the limitation of transient overcurrent, if the analytical expression of the transient current peak is... ,in If it is a set of running parameters, then the corresponding constraint inequalities are: Here, This is the maximum allowable transient current value of the equipment, such as 1.5 pu. By repeating this process for all indicators such as voltage, frequency, and power angle, a system of equations consisting of multiple inequalities is obtained, i.e., a family of constraint inequalities. Each inequality defines a hypersurface in the operating parameter space, and one side of this hypersurface is the feasible region that satisfies the specified indicator.

[0095] The system unifies and eliminates redundant constraints generated under all operating conditions. Because different disturbance conditions pose different challenges to the system, each condition generates a set of constraint inequalities. The device merges the constraint inequalities from all operating conditions into a unified constraint set, requiring parameter combinations to satisfy the performance indicators under any given condition. This unified constraint set is geometrically equivalent to the union of the feasible regions of all operating conditions. During the merging process, the system also performs redundancy analysis, eliminating loose constraints that are completely encompassed by other, more stringent constraints. For example, if the current constraint under the fault ride-through condition is... ≤1.5pu, while the constraint under load disturbance is If the value is ≤1.2 pu, and the parameter region corresponding to the latter is completely contained within the former, then the latter is redundant. After merging and eliminating, the final simplified and non-redundant set of inequalities is the voltage source characteristic constraint set. This constraint set collectively encloses a closed or semi-closed region in the multidimensional parameter space, which is the mathematical description of the parameter feasible region.

[0096] For example, the feature constraint generation module reconstructs and merges the voltage source feature constraint set based on the aforementioned derived analytical expression of transient current. First, the system obtains the voltage source characteristic maintenance condition, which stipulates that the device must effectively support the grid voltage and frequency like an ideal voltage source during transient processes, and that after experiencing severe disturbances, the grid connection point voltage and frequency must recover to the normal operating range within a specified time, while the current must not exceed the device's maximum permissible transient current value. Next, the system analyzes the voltage source characteristic maintenance condition according to three dimensions: voltage amplitude, frequency, and phase angle, obtaining specific voltage source characteristic indicators. The system then analyzes the device's maximum permissible transient current value into an indicator. Per-unit value. Subsequently, the system reconstructs the inequalities of the voltage source characteristic indicators, directly mapping the constraints of physical quantities to constraints of operating parameters, and using the analytical expression of transient current to obtain a family of constraint inequalities regarding operating parameters. Finally, the system merges the families of constraint inequalities generated under each operating condition according to the operating condition identifier and removes redundant constraints. Assuming that the system considers load disturbance and fault ride-through operating conditions during the multi-operating condition merging process, and through the analysis of merging and removing redundancy from multiple operating conditions, it is found that the constraint boundary corresponding to the load disturbance operating condition is relatively loose, while the parameter region corresponding to the more stringent constraint conditions under the fault ride-through operating condition is completely contained within the parameter region of the load disturbance operating condition. According to the principle of merging and removing redundant constraints, the strict constraints under the fault ride-through operating condition are identified as redundant and removed during the merging process. The merging principles are as follows: The union coverage principle: taking the union of all operating condition constraints ensures that the parameter combination satisfies the requirements of any operating condition; the strictness elimination principle: if the constraint boundary corresponding to the load disturbance operating condition is relatively loose, and its corresponding parameter region completely includes the parameter region corresponding to the more stringent constraint conditions under the fault ride-through operating condition, then the latter is considered redundant and eliminated; the equivalent merging principle: eliminating linearly dependent redundant constraints. After the above acquisition, analysis, reconstruction, and merging / elimination steps, the final simplified and non-redundant set of inequalities is the voltage source characteristic constraint set. This constraint set collectively encloses a closed or semi-closed region in the multi-dimensional parameter space, completing a precise mathematical description of the parameter feasible region.

[0097] S6. Perform boundary vertex search and feasibility screening on the voltage source feature constraint set to generate a parameter feasible domain vertex set;

[0098] Optionally, the set of vertexes in the feasible domain for generating parameters includes:

[0099] Solve for each constraint surface in the voltage source feature constraint set to obtain a candidate vertex set;

[0100] Each vertex in the candidate vertex set is substituted back into the voltage source feature constraint set for feasibility verification. Vertices that violate any constraint are eliminated to obtain the effective vertex set.

[0101] The set of valid vertices is clustered and duplicate points are merged according to the dimension of running parameters to generate a set of vertices in the parameter feasible domain.

[0102] Specifically, the data mapping device solves for each constraint surface in the voltage source feature constraint set to obtain a set of candidate vertices. In the dimensional parameter space, each inequality constraint defines a half-space whose boundary is a... A hyperplane of dimension 1, also known as a constraint surface. Geometrically, the vertices of the feasible region are at least 1 / 2 dimensional. The intersection points of such linearly independent constraint surfaces. Since the voltage source characteristic constraint set contains nonlinear constraints arising from the analytical expression of transient currents (such as those containing exponential and trigonometric function terms), the feasible region boundary is a curved hypersurface. The device employs a nonlinear numerical solution algorithm to search for candidate boundary vertices. For example, Newton's iteration method or sequential quadratic programming algorithm can be used to search for... Select from the constraint inequalities A system of nonlinear equations The intersection points are solved through numerical iteration; or global optimization algorithms such as genetic algorithms are used to perform boundary search in the parameter space to locate extreme points as candidate vertices. By traversing and solving different constraint combinations, the device obtains a set of candidate vertices containing all possible boundary points.

[0103] Each vertex in the candidate vertex set is then substituted back into the complete voltage source feature constraint set for feasibility verification. The intersection point obtained in the previous step only satisfies the condition that forms it. There are one constraint equality, but it is not guaranteed that it also satisfies the rest of the constraint set. There are several inequalities. Therefore, each candidate vertex must be screened. For each candidate vertex... The device will check whether it satisfies all the inequalities in the constraint set, that is:

[0104] ,

[0105] in , It is the first in the constraint set Inequalities For the voltage source characteristic constraint set, the first The constant term on the right-hand side of each inequality constraint represents the upper limit allowed by that constraint. Once discovered... If any of these constraints is violated, the vertex is deemed infeasible and eliminated. All vertices that pass this global feasibility check constitute the set of valid vertices. The feasible region vertex search and filtering results are shown in the figure below. Figure 3 As shown.

[0106] The valid vertex set is clustered and duplicate points are merged to generate the final set of vertex points in the feasible parameter domain. Due to different... A combination of constraint surfaces may calculate the same intersection point, and there may be numerically very close duplicate points in the set of valid vertices. The device is configured with a small geometric tolerance for cluster analysis of valid vertices. The geometric tolerance is determined as follows: Since the operating parameters have been standardized, the range of values ​​for each dimension of the enclosing hypercube in the parameter space is on the order of 100, and the floating-point error during the numerical solution process is approximately... Order of magnitude; to strike a balance between computational accuracy and computation time, the geometric tolerance is set to the maximum side length of the outer hypercube. The typical value is 0.01, or 1% per unit. When the Euclidean distance between two valid vertices is less than this tolerance, they are considered the same vertex, and their average coordinates are taken as the final position. This step removes computational redundancy and yields a discrete set of points that accurately describes all corners of the parametric feasible region without duplicate elements; this is the vertex set of the parametric feasible region. This vertex set forms the skeleton of the feasible region's geometry and is the sole input for subsequent boundary reconstruction.

[0107] For example, the feasible region vertex search module performs boundary vertex search and feasibility screening on the voltage source feature constraint set to reproduce the generation process of the parameter feasible region vertex set. First, the dimension of the operating parameter space is set to 2-dimensional. For the voltage source feature constraint set containing nonlinear terms, the system uses a numerical iterative algorithm to solve for the intersection points of the constraint surfaces. For example, for two nonlinear constraint surfaces... and The system iteratively calculates the parameter points that satisfy the simultaneous equations to obtain the coordinates of candidate vertices. Then, each vertex in the candidate vertex set is substituted back into the complete voltage source characteristic constraint set for feasibility verification. The verification process follows the constraint formula. If the parameter coordinates of a candidate vertex, when substituted into the third constraint, yield a function value of 3, which is greater than the upper limit of the constant term corresponding to that constraint (2), and does not satisfy the less than or equal to relationship, then it is determined to be an infeasible point that violates the constraint and is eliminated, while the valid vertices that pass the verification are retained. Finally, a geometric tolerance is set. In the set of valid vertices, it is found that two sets of vertex coordinates calculated by different constraint combinations are 2 and 2 and 2.001 and 2, respectively. The coordinate distance between the two sets is less than the set geometric tolerance, which means they are duplicate points with very similar values. Therefore, the set of valid vertices is clustered and the duplicate points are merged according to the dimension of the running parameters, and they are unified into the final vertex coordinates 2 and 2, thereby removing computational redundancy and generating a set of vertices in the parameter feasible region.

[0108] S7. The vertex set of the parameter feasible region is reconstructed and convex hull closed according to the vertex adjacency relationship to form the boundary of the parameter feasible region.

[0109] Optionally, the formation parameter feasible domain boundary includes:

[0110] Calculate the vertex adjacency relationship for the vertices in the feasible domain vertex set of the parameter;

[0111] The vertex adjacency relationships are connected sequentially to adjacent vertices according to the nearest neighbor principle to obtain the feasible region boundary polyline;

[0112] The feasible region boundary polyline is subjected to convex hull closure processing, and the internal closed-loop boundaries are identified and marked by constraint surface analysis to form the parametric feasible region boundary.

[0113] Specifically, the adjacency relationships between vertices in the feasible region vertex set are calculated. Geometrically, two vertices are adjacent if they both lie on a boundary edge of the feasible region. In engineering implementation, the device checks any two vertices... and If it exists The same constraint inequalities are in and If equality holds at both points, then these two vertices lie on this... A boundary edge is formed by the intersection of several constraint surfaces. The device systematically examines all vertex pairs, records all adjacency relationships that satisfy this condition, and forms an adjacency matrix or adjacency list describing vertex connectivity. This data structure serves as the blueprint for constructing the feasible region boundary skeleton.

[0114] Based on the calculated vertex adjacency relationships, adjacent vertices are connected sequentially according to the nearest neighbor principle to form a feasible region boundary polyline. Starting from any vertex, the device queries its adjacency relationships, selects an unvisited adjacent vertex, and connects it to form a line segment. Then, with the new vertex as the current node, this process is repeated until the starting vertex is returned, forming a closed polygonal loop. If the feasible region is a polyhedron, this process will generate multiple such faces. By traversing all vertices and adjacency relationships, the system connects discrete vertices into a grid-like structure composed of line segments, i.e., the preliminary feasible region boundary polyline. The feasible region boundary polyline is then subjected to boundary closure processing. For local gaps caused by numerical calculation errors, a convex hull algorithm is used to repair them to form a closed convex external boundary; at the same time, internal closed-loop boundaries are identified and marked through constraint surface analysis to form the parametric feasible region boundary. Due to calculation errors or the non-convex nature of the feasible region itself, the initially formed boundary polyline may have gaps, or there may be holes inside the feasible region caused by not satisfying certain nonlinear constraints. The device employs standard computational geometry algorithms, such as Quickhull or Gift wrapping, to calculate the convex hull of all vertices. Taking the algorithm as an example, its calculation process is as follows: Select the point with the smallest x-coordinate. Starting from vertex A, add it to the convex hull vertex list. Let the current vertex be A. Select from the remaining vertices As a temporary candidate point, traverse all other vertices. Calculate the cross product:

[0115] ,

[0116] In the formula This represents the x-coordinate of a vertex in a two-dimensional coordinate system. Represents the ordinate of a vertex in a two-dimensional coordinate system. <0 indicates lie in → The clockwise direction of the connecting line, that is Further outwards, making = After the traversal is complete, This is the next vertex of the convex hull. Add to the convex hull list, let = Repeat the above process until the current vertex returns to the starting point. At this point, the convex hull is closed.

[0117] A convex hull is a minimal convex polygon or polyhedron containing all vertices. This operation ensures the integrity and closure of the outer boundary of the feasible region. For potential internal holes, the device identifies them by analyzing the connectivity between vertices and their relationship to the constraint surfaces. Once a hole boundary is identified, it is marked and preserved as an internal boundary. After convex hull closure and internal hole processing, a watertight manifold surface that accurately describes the geometry of the parametric feasible region is obtained; this is the boundary of the parametric feasible region. This boundary provides a solid geometric foundation for subsequent quantitative analysis.

[0118] For example, the feasible region boundary reconstruction module reconstructs the vertex set of the parameter feasible region according to vertex adjacency relationships and performs convex hull closure to reproduce the formation process of the parameter feasible region boundary. First, the dimension of the running parameter space is set. At this point, the feasible vertex set of the input parameters includes external vertices. The coordinates are 0 and 0. The coordinates are 2 and 0, plus a small deviation of 2.001 and 0 due to calculation error. The coordinates are 2 and 2, Coordinates 0 and 2, and the vertex of the internal hole. The coordinates are 0.5 and 0.5. The coordinates are 1.5 and 0.5. The coordinates are 1.5 and 1.5, which are processing points. The system calculates the adjacency relationship between vertices in the vertex set. For any two vertices, if there exists... If two identical constraint inequalities, 2-1=1, are simultaneously equal, then the pair is considered adjacent. Through traversal calculations, it was found that... and Together on the constraint surface The system checks if the condition 2-1=1 is met, then determines if the vertices are adjacent and records this relationship in the adjacency matrix. Next, the system connects adjacent vertices sequentially according to the nearest neighbor principle based on the adjacency matrix. Starting from the nearest neighbor, line segment loops are generated sequentially, connecting discrete vertices into an external structure composed of multiple line segments and non-closed loops composed of internal points, resulting in a preliminary feasible region boundary polyline. Finally, the system performs convex hull closure processing on the generated feasible region boundary polyline and identifies and marks the internal closed-loop boundaries through constraint surface analysis. The Quickhull algorithm is used to calculate the convex hull of all vertices, generating a minimal convex polygon containing all vertices as the external boundary to eliminate gaps caused by off-center points and ensure the closure of the external boundary. Simultaneously, by analyzing the connectivity between vertices and their relationship with constraint surfaces, the system identifies the internal closed-loop boundaries formed by... and The nonlinear non-satisfied region at the axial intersection is marked and preserved as an internal closed-loop boundary, ultimately forming a watertight and closed parametric feasible domain boundary.

[0119] S8. Perform volume measurement and operational margin quantization calculation on the feasible domain boundary of the parameter to obtain the feasible domain quantization result, and then output the feasible domain quantization result after mapping it to the characteristic operating boundary of the full-condition voltage source.

[0120] Optionally, the step of performing volume measurement and operational margin quantization calculations on the feasible region boundary of the parameters to obtain the feasible region quantization result, and then mapping the feasible region quantization result to the characteristic operational boundary of the voltage source under all operating conditions, includes the following output:

[0121] The volume of the feasible region boundary of the parameter is measured by random sampling integration to obtain the feasible region capacity index.

[0122] Calculate the target normal distance from the current operating point of the grid-type energy storage device to the boundary of the feasible region of the parameters, and obtain the feasible region margin index.

[0123] The feasible region capacity index and the feasible region margin index are fused according to the working condition weight to obtain the feasible region quantification result.

[0124] The quantization results of the feasible region are organized into an operational boundary lookup table according to the operational parameter dimension, which serves as the operational boundary output of the full-condition voltage source characteristics.

[0125] Specifically, random sampling integration, i.e., the Monte Carlo integration method, is used to measure the volume enclosed by the feasible region boundary of the parameters. The calculation formula is as follows:

[0126] ,

[0127] Represents the volume of the feasible region; This represents the volume of the enclosing hypercube containing the feasible region; This represents the number of random sampling points that fall within the feasible region; This represents the total number of random sampling points. The system generates a large number of uniformly distributed random parameter points within a hypercube containing the feasible region. By statistically analyzing the ratio of the number of points falling within the feasible region to the total number of sampling points, and then multiplying this ratio by the volume of the hypercube, the volume of the feasible region can be estimated. This volume value is defined as the feasible region capacity index. This index intuitively reflects the size of the parameter adjustment space of the system under all constraints. The feasible region margin index is obtained by calculating the target normal distance from the current operating point of the grid-type energy storage device to the boundary of the parameter feasible region. The target normal distance refers to the shortest Euclidean distance from the current operating point of the grid-type energy storage device to the boundary of the feasible region in the operating parameter space, and this distance vector is consistent with the outward normal direction of the boundary at the nearest point. This distance is obtained by solving a geometric optimization problem with outward normal direction constraints, characterizing the geometric safety margin of the current operating point in the multidimensional parameter space, which is different from the stability margin based on eigenvalues ​​or damping ratios. The current operating point is a point composed of real-time collected operating parameters. The target normal distance refers to the shortest distance from the point to the feasible region boundary, typically along the normal direction of the boundary. This distance can be calculated by solving a constrained optimization problem; its magnitude represents the safe distance from the current operating state to triggering an unstable boundary. A larger margin indicates a more stable system. The obtained feasible region capacity index and feasible region margin index are fused according to preset operating condition weights to obtain the final feasible region quantification result. The operating condition weights reflect the differences in the importance of different operating scenarios to system safety. The fusion process can be expressed by the following formula:

[0128] ,

[0129] In this formula, This represents the final feasible domain quantification result. For the volume of the outer hypercube, It is the shortest normal distance. This is the maximum inner diameter in the parameter space. and These are the weighting coefficients corresponding to capacity and margin, respectively. They are dynamically adjusted based on the risk assessment of the current operating conditions, and their sum is 1.

[0130] The quantification results of this feasible region are organized according to the dimensions of operating parameters to generate an operating boundary lookup table. This table, indexed by key operating parameters, provides feasible region boundary information and margin levels under different parameter combinations, thus constituting the characteristic operating boundary of the voltage source under all operating conditions. This is output to scheduling or maintenance personnel through a human-machine interface or data interface. The final output is a structured characteristic operating boundary of the voltage source under all operating conditions. This output is not a single numerical value, but a comprehensive decision support information package, typically presented in the form of a lookup table or visualization. It includes an overall assessment of the flexibility of system parameter adjustments (feasible region capacity index), a real-time assessment of the safety of the current operating point (feasible region margin index), and a tool that allows users to quickly query the safety boundary under any parameter combination. This quantification result transforms complex system stability analysis into clear and intuitive operational guidance, enabling operators to optimize parameters and predict risks based on accurate data. A multi-dimensional comparison chart of the feasible region quantification results is shown below. Figure 4 As shown.

[0131] For example, the quantization evaluation and output module performs volume measurement and operational margin quantization calculations on the parameter feasible region boundary to reproduce the output process of the characteristic operating boundary of the voltage source under full operating conditions. First, the system uses a random sampling integration method to perform volume measurement on the parameter feasible region boundary, setting the volume of the hypercube space containing this feasible region as... =10, evenly distributed within this space =10000 random parameter sampling points, statistically, there are a total of =6500 sampling points fall within the feasible region boundary of the parameters. Calculate their proportion and multiply by the spatial volume to obtain the feasible region volume. Simultaneously calculate the volume fill rate. =0.65, used as the normalized feasible region capacity index. Next, the system collects real-time data on the current operating point of the grid-type energy storage device, calculates the shortest target normal distance from the current operating point to the boundary of the parameter feasible region by solving a constrained optimization problem, and sets the maximum inner diameter of the parameter space as... =5, and the normal distance to the target is calculated to be 1.2 per unit, thus obtaining the normalized feasible region margin index. Subsequently, the system dynamically adjusts the weighting coefficients based on the risk assessment results of the current power grid operating conditions, and sets the weighting coefficients for the corresponding capacities. The corresponding margin weighting coefficient The sum of the two weighting coefficients must be equal to 1. The feasible region capacity index and feasible region margin index are then substituted into the fusion formula according to the load condition weights for calculation, yielding the final feasible region quantification result. Finally, the system quantifies the obtained feasible region results. Organize and associate key operating parameters to generate corresponding structured operating boundary lookup tables, and output them to dispatchers through data interfaces as the characteristic operating boundaries of the voltage source under all operating conditions. This completes the global and accurate quantitative characterization of the system parameter adjustment space and safety status.

[0132] Based on the same inventive concept, such as Figure 5 As shown, the present invention also provides a voltage source feature feasible region quantization system for a data mapping device, the system comprising:

[0133] The parameter acquisition and preprocessing module is used to acquire electrical topology parameters and control loop parameters, and to merge and integrate the acquired results according to the category of physical quantities to obtain a set of operating parameters.

[0134] The dynamic characteristic hierarchical module is used to hierarchically divide the set of operating parameters to obtain the slow dynamic quantity of the control loop and the fast dynamic quantity of the network electromagnetic transient.

[0135] The decoupled dynamics modeling module is used to establish dynamic sub-equations based on the slow dynamics of the control loop and the fast electromagnetic transients of the network, respectively, and to perform slow and fast variable algebraization to reduce the order of the dynamic sub-equations to construct a decoupled dynamics model.

[0136] The transient current analysis module is used to load all types of disturbance excitation based on the decoupled dynamic model, solve the transient response trajectory of each condition, and extract the explicit expression containing the operating parameters through integral transformation and algebraic elimination to obtain the analytical expression of the transient current.

[0137] The feature constraint generation module is used to perform constraint mapping on the analytical expression of the transient current, merge constraints and remove redundancy, and generate a voltage source feature constraint set.

[0138] The feasible domain vertex search module is used to perform boundary vertex search and feasibility screening on the voltage source feature constraint set to generate a parameter feasible domain vertex set.

[0139] The feasible region boundary reconstruction module is used to reconstruct the vertex set of the parameter feasible region according to the vertex adjacency relationship and close the convex hull to form the parameter feasible region boundary;

[0140] The quantization evaluation and output module is used to perform volume measurement and operational margin quantization calculations on the feasible domain boundary of the parameters, obtain the feasible domain quantization result, and output the feasible domain quantization result after mapping it to the characteristic operating boundary of the voltage source under all operating conditions.

[0141] It should be noted that the electrical connections between the various units described above do not necessarily represent direct or indirect connections. Any indirect connection method can be applied to the embodiments of the present invention as long as it achieves the purpose of the present invention. The above descriptions are merely exemplary embodiments of the present invention and should not be construed as limiting the scope of the present invention.

[0142] All equivalent changes and modifications made in accordance with the teachings of this invention are still within the scope of this invention. Those skilled in the art will readily conceive of other embodiments of this invention upon considering the specification and the disclosure of practical truth. This application is intended to cover any variations, uses, or adaptations of this invention that follow the general principles of this invention and include common knowledge or conventional techniques in the art not described herein.

Claims

1. A method for quantizing the feasible region of voltage source characteristics in a data mapping device, characterized in that, include: Electrical topology parameters and control loop parameters are collected, and the collected results are merged and integrated according to the category of physical quantities to obtain a set of operating parameters; The set of operating parameters is divided into hierarchical components to obtain the slow dynamics of the control loop and the fast dynamics of the network electromagnetic transients. Based on the slow dynamics of the control loop and the fast transient dynamics of the network electromagnetic system, dynamic sub-equations are established respectively. The dynamic sub-equations are then subjected to slow-fast variable algebraization to reduce their order, and a decoupled dynamic model is constructed. Based on the decoupled dynamic model, all types of disturbance excitation are applied, the transient response trajectories of each condition are solved, and after integral transformation and algebraic elimination, explicit expressions containing operating parameters are extracted to obtain the analytical expression of transient current. The transient current analytical expression is constrained and mapped, constraints are merged and redundancy is eliminated to generate a voltage source feature constraint set; Boundary vertex search and feasibility screening are performed on the voltage source feature constraint set to generate a parameter feasible domain vertex set; The vertex set of the parameter feasible region is reconstructed and its convex hull is closed according to the vertex adjacency relationship to form the boundary of the parameter feasible region; The feasible region boundary of the parameters is quantized by volume measurement and operating margin to obtain the feasible region quantization result, and the feasible region quantization result is mapped to the characteristic operating boundary of the voltage source under all operating conditions and then output.

2. The voltage source feature feasible region quantization method for a data mapping device according to claim 1, characterized in that, The collected electrical topology parameters and control loop parameters are then merged and integrated according to physical quantity categories to obtain a set of operating parameters, including: Read the bus voltage phasor, branch current phasor, and switch position, and arrange them according to the topology node number to obtain the electrical topology parameters; Extract the droop coefficient, virtual inertia coefficient, and voltage outer loop bandwidth, and label them according to the control level to obtain the control loop parameters; The electrical topology parameters and the control loop parameters are merged and integrated according to physical quantity categories and then standardized to obtain a set of operating parameters.

3. The voltage source feature feasible region quantization method for a data mapping device according to claim 1, characterized in that, The hierarchical division of the operating parameter set to obtain the slow dynamic quantities of the control loop and the fast dynamic quantities of the network electromagnetic transients includes: Characteristic frequency bands are identified for each physical quantity in the set of operating parameters to obtain parameter subsequences with frequency band identifiers; Based on the frequency band identifier, the parameter subsequence is decomposed into low-frequency components and high-frequency components to obtain a hierarchical parameter sequence; The low-frequency components in the hierarchical parameter sequence are aggregated over time according to the control loop response window to obtain the slow dynamics of the control loop. The high-frequency components in the hierarchical parameter sequence are extracted according to the electromagnetic transient time window to obtain the network electromagnetic transient fast dynamic quantity.

4. The voltage source feature feasible region quantization method for a data mapping device according to claim 1, characterized in that, The construction of the decoupled dynamic model includes: Based on the slow dynamics of the control loop, a control loop state equation containing a droop term and a virtual inertia term is constructed, thus obtaining the control loop dynamic sub-equation. Based on the network electromagnetic transient fast dynamics, a transient network equation including the relationship between inductor-capacitor branches and node admittance is constructed, and the network dynamics sub-equation is obtained; The control loop dynamics sub-equations and the network dynamics sub-equations are reduced in order and subjected to boundary matching by slow-fast variable algebraization based on the time scale difference, resulting in a decoupled dynamics model.

5. The voltage source feature feasible region quantization method for a data mapping device according to claim 1, characterized in that, The process involves loading excitations for all types of disturbances based on the decoupled dynamic model, solving for the transient response trajectories under different conditions, and then extracting explicit expressions containing operating parameters through integral transformation and algebraic elimination to obtain the analytical expression of the transient current, including: Set a set of disturbance conditions including startup, grid connection, load disturbance, and fault ride-through; The set of disturbance conditions is loaded into the decoupled dynamic model as excitation input item by item to obtain the transient response trajectory of each condition. The transient response trajectories under the different operating conditions are sequentially subjected to time-domain integral transformation and algebraic elimination, and explicit expressions for the operating parameters are extracted to obtain the analytical expression of the transient current.

6. The voltage source feature feasible region quantization method for a data mapping device according to claim 1, characterized in that, The set of characteristic constraints for the generated voltage source includes: Obtain the voltage source characteristic sustaining conditions; The voltage source characteristic maintenance conditions are analyzed according to three dimensions: voltage amplitude, frequency, and phase angle, to obtain the voltage source characteristic index. By combining the analytical expression of the transient current with the inequality reconstruction of the voltage source characteristic index, a family of constraint inequalities about the operating parameters is obtained; The family of constraint inequalities is combined according to the operating condition identifier and redundant constraints are removed to generate a voltage source characteristic constraint set.

7. The voltage source feature feasible region quantization method for a data mapping device according to claim 1, characterized in that, The feasible vertex set for generating parameters includes: Solve for each constraint surface in the voltage source feature constraint set to obtain a candidate vertex set; Each vertex in the candidate vertex set is substituted back into the voltage source feature constraint set for feasibility verification. Vertices that violate any constraint are eliminated to obtain the effective vertex set. The set of valid vertices is clustered and duplicate points are merged according to the dimension of running parameters to generate a set of vertices in the parameter feasible domain.

8. The voltage source feature feasible region quantization method for a data mapping device according to claim 1, characterized in that, The feasible domain boundary for the formation parameters includes: Calculate the vertex adjacency relationship for the vertices in the feasible domain vertex set of the parameter; The vertex adjacency relationships are connected sequentially to adjacent vertices according to the nearest neighbor principle to obtain the feasible region boundary polyline; The feasible region boundary polyline is subjected to convex hull closure processing, and the internal closed-loop boundaries are identified and marked by constraint surface analysis to form the parametric feasible region boundary.

9. The voltage source feature feasible region quantization method for a data mapping device according to claim 1, characterized in that, The step involves quantizing the volumetric measure and operational margin of the feasible region boundary of the parameters to obtain the feasible region quantization result. The output, after mapping the feasible region quantization result to the characteristic operational boundary of the full-condition voltage source, includes: The volume of the feasible region boundary of the parameter is measured by random sampling integration to obtain the feasible region capacity index. Calculate the target normal distance from the current operating point of the grid-type energy storage device to the boundary of the feasible region of the parameters, and obtain the feasible region margin index. The feasible region capacity index and the feasible region margin index are fused according to the working condition weight to obtain the feasible region quantification result. The quantization results of the feasible region are organized into an operational boundary lookup table according to the operational parameter dimension, which serves as the operational boundary output of the full-condition voltage source characteristics.

10. A voltage source feature feasible region quantization system for a data mapping device, applied to the voltage source feature feasible region quantization method for a data mapping device as described in any one of claims 1-9, characterized in that, The system includes: The parameter acquisition and preprocessing module is used to acquire electrical topology parameters and control loop parameters, and to merge and integrate the acquired results according to the category of physical quantities to obtain a set of operating parameters. The dynamic characteristic hierarchical module is used to hierarchically divide the set of operating parameters to obtain the slow dynamic quantity of the control loop and the fast dynamic quantity of the network electromagnetic transient. The decoupled dynamics modeling module is used to establish dynamic sub-equations based on the slow dynamics of the control loop and the fast electromagnetic transients of the network, respectively, and to perform slow and fast variable algebraization to reduce the order of the dynamic sub-equations to construct a decoupled dynamics model. The transient current analysis module is used to load all types of disturbance excitation based on the decoupled dynamic model, solve the transient response trajectory of each condition, and extract the explicit expression containing the operating parameters through integral transformation and algebraic elimination to obtain the analytical expression of the transient current. The feature constraint generation module is used to perform constraint mapping on the analytical expression of the transient current, merge constraints and remove redundancy, and generate a voltage source feature constraint set. The feasible domain vertex search module is used to perform boundary vertex search and feasibility screening on the voltage source feature constraint set to generate a parameter feasible domain vertex set. The feasible region boundary reconstruction module is used to reconstruct the vertex set of the parameter feasible region according to the vertex adjacency relationship and close the convex hull to form the parameter feasible region boundary; The quantization evaluation and output module is used to perform volume measurement and operational margin quantization calculations on the feasible domain boundary of the parameters, obtain the feasible domain quantization result, and output the feasible domain quantization result after mapping it to the characteristic operating boundary of the voltage source under all operating conditions.

Images