A simulation method for dynamic characteristics of new energy transformer substation under main grid fault

Abstract

The application provides a simulation method for dynamic characteristics of a new energy transformer area under main grid failure, and relates to the technical field of power system stability analysis. The method comprises the following steps: constructing an output power model of the new energy transformer area during fault ride-through; based on the output power model, calculating the output power of each stage according to the removal sequence of each unit, and constructing a power response curve; selecting a reference unit, and building a single-machine grid-connected model on an electromagnetic transient simulation platform; simulating each voltage mutation scene with the reference unit as the object, and extracting a reference transient response waveform; based on the reference transient response waveform, calculating a transient correction amount according to the ratio of the reference capacity of the single-machine grid-connected model to the rated capacity; and superimposing the transient correction amount to the power response curve to obtain a dynamic power response curve of the new energy transformer area. The method can accurately depict the whole process dynamic power characteristics of the high-proportion new energy transformer area under the voltage drop scene.

Description

Technical Field

[0001] This invention relates to the field of power system stability analysis technology, and in particular to a simulation method for the dynamic characteristics of new energy distribution areas under main grid faults. Background Technology

[0002] In power systems characterized by photovoltaic, wind, and energy storage, when a short-circuit fault on the main grid causes a voltage drop in renewable energy distribution areas, each generating unit will enter a fault ride-through state according to the requirements of the corresponding national standards for renewable energy grid connection, injecting dynamic reactive current into the grid to support voltage recovery. Disconnection from the grid is only permitted after the fault duration exceeds the specified tolerance curve. Due to differences in voltage distribution caused by line impedance, the voltage drop depth varies among different generating unit nodes. Units closer to transformers may enter low-voltage ride-through mode first, while units further away may temporarily maintain normal operation. As the fault continues, some generating units disconnect from the grid sequentially due to reaching the fault ride-through time limit, further triggering a redistribution of power flow and chain reactions in the power system. This multi-stage, multi-event dynamic process results in complex time-varying power characteristics of renewable energy distribution areas as seen from the main grid during fault periods. However, to ensure the safe and stable operation of power systems with a high proportion of renewable energy, simulation tools capable of quickly and accurately acquiring the dynamic power response of such distribution areas under various voltage drop conditions are urgently needed to support power system-level fault scanning, relay protection setting, and stability control strategy formulation.

[0003] In terms of simulation methods for the dynamic characteristics of voltage dips in high-proportion renewable energy distribution areas, existing technologies include two categories: detailed electromagnetic transient simulation and simplified equivalent modeling simulation. Detailed electromagnetic transient simulation methods establish complete circuit and control models for each generator unit, collector line, and transformer, and use small-step numerical integration for solution, accurately reproducing the waveforms of voltage and current at each node during a fault. However, for distribution areas with dozens or more generator units, this method faces bottlenecks due to high model order and slow simulation speed; a single fault scenario simulation can take several minutes to tens of minutes, making it difficult to meet the timeliness requirements of rapid scanning of a large number of scenarios in system-level analysis. On the other hand, simplified equivalent modeling simulation methods attempt to reduce computational load by aggregating the entire distribution area or some of its generator units into a single equivalent machine. Among them, the single-unit aggregation method treats all power generation units as a single model, ignoring the differences in voltage distribution within the distribution area, and cannot reflect the mixed state in reality where "some units enter fault ride-through while others are still operating normally"; although the multi-unit aggregation method considers voltage distribution factors, existing studies mostly focus on power distribution calculations during fault steady state, failing to incorporate fault ride-through time constraints and sequential grid disconnection dynamic processes into the simulation framework, resulting in the inability to generate complete time-domain power response curves.

[0004] Furthermore, both of the aforementioned methods neglect the transient response of the inverter's internal control when handling power mode switching of the generator unit (such as switching from constant power mode to fault ride-through mode or vice versa). While detailed electromagnetic transient simulation can characterize this transient, its computational cost is too high; simplified equivalent methods generally assume that the power mode switching is instantaneous, ignoring the transient adjustment process of several milliseconds to tens of milliseconds caused by the inverter's internal PI controller and output filter. In reality, the transient adjustment process cannot be ignored in the short-timescale system dynamic analysis after a fault, and the power overshoot or damped oscillations it causes will affect the action criteria of the protection device and the stability margin assessment.

[0005] In short, existing methods cannot accurately characterize the dynamic power characteristics of high-proportion renewable energy distribution areas throughout the entire process under voltage drop scenarios. Summary of the Invention

[0006] To address the aforementioned technical problems in existing technologies, this invention aims to provide a simulation method for the dynamic characteristics of renewable energy distribution areas under main grid faults, so as to accurately characterize the full-process dynamic power characteristics of high-proportion renewable energy distribution areas under voltage dip scenarios. Specifically, the technical solution is as follows:

[0007] A simulation method for the dynamic characteristics of renewable energy distribution areas under main grid failure includes the following steps:

[0008] Step S1: Set the low-voltage side node of the transformer in the distribution area as the initial boundary point of the node voltage and update iteratively to construct the output power model of the new energy distribution area during fault ride-through.

[0009] Step S2: Determine the disconnection sequence based on the continuous non-disconnection operation time of each unit; based on the output power model, calculate the output reactive power and output active power of the new energy distribution area at each stage according to the disconnection sequence of each unit, and construct the power response curve;

[0010] Step S3: Select the unit closest to the transformer in the new energy distribution area as the benchmark unit, and build a single-unit grid-connected model on the electromagnetic transient simulation platform;

[0011] Step S4: Using the benchmark unit as the object, perform simulations of various voltage change scenarios and extract the corresponding benchmark transient response waveforms;

[0012] Step S5: Based on the reference transient response waveform, calculate the transient correction amount of output active power and output reactive power in each unit according to the ratio of the reference capacity to the rated capacity of the single-unit grid-connected model.

[0013] Step S6: Superimpose the transient correction onto the power response curve to obtain the dynamic power response curve of the new energy distribution area.

[0014] Preferably, step S1 includes:

[0015] Step S11: Set all units to be in normal operation after a voltage drop occurs, and use the low-voltage side of the transformer in the distribution area as the initial dividing point. ;

[0016] Step S12: Includes steps S121 to S124;

[0017] Step S121: Set the initial amplitude of each node voltage to the node voltage on the low-voltage side of the transformer in the distribution area. ;

[0018] Step S122: Calculate the output reactive power and output active power of the corresponding unit based on the voltage amplitude of each node and the operating status of the corresponding unit;

[0019] Step S123: Calculate the total output reactive power and total output active power of each downstream unit at each node;

[0020] Step S124: Update the voltage of each node based on the total output reactive power and the total output active power, and calculate the difference in the voltage amplitude of the corresponding node before and after the update;

[0021] If all differences are less than the preset value, continue; otherwise, return to step S12.

[0022] Step S13: Update the boundary points according to the per-unit values ​​of the updated node voltages;

[0023] If the updated boundary point is equal to the original boundary point, continue; otherwise, return to step S12.

[0024] Step S14: Based on the output reactive power and output active power of each unit, calculate the total reactive power and total active power at the transformer in the distribution area to obtain the output power model.

[0025] Preferably, step S122 includes:

[0026] Determine the operating status of each unit: if node Less than the current dividing point Then the first The unit is in a low-voltage ride-through state; otherwise, the first The unit is in normal operating condition;

[0027] Calculate the first The output reactive power and output active power of each generating unit include:

[0028] (1) If the first The generator unit is in a low voltage ride-through state; calculate the output reactive power. and output active power The formula is as follows:

[0029] ;

[0030] ;

[0031] ;

[0032] ;

[0033] ;

[0034] ;

[0035] ;

[0036] In the formula, For node sequence number, , For the number of nodes, For nodes The node voltage amplitude, For the first The output active current of the inverter in the Taiwanese unit. For the first The output reactive current of the inverter in the Taiwanese unit For the first Reference value of the output reactive current of the inverter in the Taiwanese unit. For the first The maximum allowable current of the inverter in the Taiwanese unit. For the first The output active current of the inverter in the Taiwanese unit before the inverter failure. For the first The initial current of the generator set, This is the dynamic reactive current proportionality coefficient. For nodes The per-unit value of the node voltage. For the first The rated current of the unit For the first Rated active power output of the generator set For the first The node voltage amplitude before the inverter failure in the Taiwanese unit. This is the nominal line voltage;

[0037] (2) If the first If the generator set is in normal operating condition, then the first... The output reactive power of the inverter in the Taiwanese unit For the first Rated reactive power output of the unit Output active power For the first Rated active power output of the Taiwanese generator set ;

[0038] The formula for step S123 is as follows:

[0039] ;

[0040] ;

[0041] In the formula, For the first The total active power output of the generator unit and its downstream units. For the first The total reactive power output of the generator unit and its downstream units. For unit serial number, ;

[0042] The voltage at each node is updated based on the total output reactive power and the total output active power, using the following formula:

[0043] ;

[0044] In the formula, The node voltage is denoted by . For nodes With nodes The line resistance between them For nodes With nodes Line reactance between;

[0045] The step of updating the boundary point based on the per-unit value of each node voltage includes:

[0046] per unit value The first node serves as the new dividing point. If the per-unit value corresponding to all nodes Then let the dividing point ;

[0047] The formula for step S14 is as follows:

[0048] ;

[0049] ;

[0050] ;

[0051] ;

[0052] ;

[0053] In the formula, This refers to the total active power at the transformer station in the distribution area. This represents the total reactive power at the transformer substation. For nodes With nodes The three-phase active power loss of the line between them. For nodes With nodes The three-phase active power loss of the line between them. Line impedance The effective value of the current flowing through it.

[0054] Preferably, step S2 includes:

[0055] Step S21: Set the voltage drop on the low-voltage side of the transformer in the distribution area when... Based on the node voltage of the low-voltage side of the transformer in the distribution area Calculate the node voltage, output reactive power and output active power of each grid connection point before the first unit is disconnected according to step S122.

[0056] For all units in the low voltage ride-through state, from Accumulate downtime during fault operation;

[0057] Step S22: When the new energy distribution area reaches the minimum value of the continuous non-disconnection operation time of each node, the corresponding unit is disconnected;

[0058] Step S23: After the generator unit is disconnected, update the node voltage and non-disconnection continuous operation time of the remaining nodes;

[0059] Step S24: Repeat steps S22 to S23 until no unit is in a low voltage ride-through state or the fault is cleared;

[0060] The power response curve is constructed using the following formula:

[0061] ;

[0062] In the formula, This is the steady-state active power curve of the new energy distribution area. The steady-state reactive power curve of the new energy distribution area. This is the first time the steady-state active power of the new energy distribution area has been cut off. This is the first time the steady-state reactive power of the previous renewable energy distribution area has been cut off. This represents the steady-state active power of the renewable energy distribution area after the first disconnection. This represents the steady-state reactive power of the renewable energy distribution area after the first disconnection. The time of the first excision. This refers to the time of the second excision.

[0063] Furthermore, it also includes a method for updating the continuous online running time, as shown in the following formula:

[0064] ;

[0065] In the formula, For nodes The continuous operating time without disconnecting from the network. Per-unit value of node voltage A mapping function to the allowed time.

[0066] Preferably, step S4 includes:

[0067] Scenario A: For units that enter the low-voltage ride-through state synchronously with a voltage dip:

[0068] In the single-machine grid-connected model, the node voltage is set to the normal operating value of node 1 before the fault, and the inverter operates stably at the rated power.

[0069] Let the node voltage be at When the voltage drops to the node voltage of node 1 after the voltage drop of the first generating unit, When the inverter returns to the normal operating value of node 1, the time-domain response curves of the output active power and output reactive power of the inverter are recorded as the corresponding reference transient response waveforms.

[0070] Scenario B: For units that are still operating normally when the voltage drops:

[0071] Update the boundary point according to step S13;

[0072] In the single-unit grid-connected model, the node voltage is set to the normal operating value at the boundary point before the fault, and the inverter operates stably at the rated power.

[0073] Let the node voltage be at When the voltage drops to the node voltage at the breakpoint after the corresponding unit experiences a voltage drop, When the inverter returns to its normal operating value at the boundary point, record the time-domain response curves of the inverter's output active power and output reactive power as the corresponding reference transient response waveforms.

[0074] Scenario C: For units that are still operating normally when the voltage drops, but enter a low-voltage ride-through state during the voltage drop:

[0075] In the stand-alone grid-connected model, the simulations of scenario B and scenario A are executed sequentially, and the time-domain response curves of the inverter's output active power and output reactive power are recorded as the corresponding reference transient response waveforms.

[0076] Preferably, the formula for step S5 is as follows:

[0077] ;

[0078] ;

[0079] ;

[0080] In the formula, For the first The transient correction of the active power output of the generator unit. For the first The transient correction amount of the reactive power output of the generator unit. For the first The capacity ratio factor of the Taiwanese unit. To output the time-domain response curve of active power, This represents the steady-state active power of the stand-alone grid-connected model during voltage dips. The time-domain response curve for output reactive power. This represents the steady-state reactive power of a single-unit grid-connected model during a voltage dip. For the first The rated capacity of the unit This is the base capacity corresponding to the base unit;

[0081] The formula for step 6 is as follows:

[0082] ;

[0083] ;

[0084] In the formula, The dynamic power response curve for the active power output of the new energy distribution area. The dynamic power response curve for reactive power output from the new energy distribution area. For the occurrence time of each voltage surge event, This refers to the set of generating units involved in voltage surge events. When a voltage surge event occurs, the first The transient correction of the active power output of the generator unit. When a voltage surge event occurs, the first The transient correction of reactive power output by the generator unit; among which, voltage surge events include voltage sag, unit disconnection and voltage recovery.

[0085] Compared to existing technologies, the technical solution provided by this invention effectively represents the initial mixed-operation steady state considering the voltage distribution differences within the renewable energy distribution area and the dynamic load voltage dependence characteristics by constructing an output power model. It accurately depicts the event sequence of step-by-step disconnection and operating state switching of each unit by constructing power response curves, and incorporates the transient processes within the inverter into the simulation model through a single-unit grid-connected model, accurately reflecting the transient responses generated by unit operating state switching. In short, the technical solution provided by this invention can accurately depict the full-process dynamic power characteristics of a high-proportion renewable energy distribution area under voltage drop scenarios. Attached Figure Description

[0086] Figure 1 This is a flowchart illustrating a simulation method in one embodiment of the present invention.

[0087] Figure 2 This is a schematic diagram of the process for constructing a model of the output power of a new energy transformer area during fault ride-through, according to one embodiment of the present invention.

[0088] Figure 3 This is a schematic diagram of the process for constructing a power response curve in one embodiment of the present invention.

[0089] Figure 4 This is a schematic diagram illustrating the process of obtaining the dynamic power response curve based on a single-machine grid-connected model in one embodiment of the present invention.

[0090] Figure 5 This is a time-limit curve of low voltage ride-through in one embodiment of the present invention.

[0091] Figure 6 This is a diagram showing the voltage distribution at nodes before and after a voltage drop in this invention.

[0092] Figure 7 This is a schematic diagram of the voltage drop threshold point in one embodiment of the present invention.

[0093] Figure 8 This is a schematic diagram of the inner layer power flow iteration and outer layer iteration in one embodiment of the present invention.

[0094] Figure 9 This is a diagram showing the output distribution of active and reactive power of each unit before and after a voltage drop in one embodiment of the present invention.

[0095] Figure 10 This is a power response curve of a new energy distribution area in one embodiment of the present invention.

[0096] Figure 11 This is a schematic diagram of the transient response extraction results of a single-machine grid-connected model in one embodiment of the present invention.

[0097] Figure 12 This is a power response curve of the new energy distribution area after correction in one embodiment of the present invention.

[0098] Figure 13 This is the dynamic power response curve of a new energy distribution area in one embodiment of the present invention. Detailed Implementation

[0099] The technical solutions provided by the present invention will be further described in detail below with reference to the embodiments and accompanying drawings.

[0100] Example 1:

[0101] like Figure 1 As shown, in order to solve the following problems in the dynamic simulation of high-proportion renewable energy distribution areas in the existing technology: the voltage distribution differences caused by line impedance within the distribution area are difficult to take into account, the dynamic process of sequential grid disconnection of units under fault ride-through time constraints cannot be effectively simulated, and the millisecond-level transient response generated by the internal control loop of the inverter is generally ignored, this invention proposes a rapid simulation method for voltage drop in high-proportion renewable energy distribution areas that takes into account sequential grid disconnection and inverter transients.

[0102] First, an initial threshold point for node voltage is set and iteratively updated to construct the output power model of the renewable energy distribution area during fault ride-through. Based on the magnitude power flow recursive equation ignoring voltage phase angle, and combined with the low-voltage ride-through power model of renewable energy units at the inverter interface, the steady-state distribution of node voltages of each unit after voltage drop is calculated through iterative updates of the node voltage threshold point and double-layer iterative solution. This determines the mixed operating state of "some units entering low-voltage ride-through and some units maintaining normal operation" caused by voltage distribution differences, obtains the initial steady-state power response of the distribution area before the first unit is disconnected, and establishes the output power model.

[0103] Then, based on the output power model, the output reactive power and output active power of the new energy distribution area at each stage are calculated according to the disconnection sequence of each unit, and the power response curve is constructed. Starting from the initial steady-state power response, the fault ride-through time constraint specified by the grid connection standard is introduced to establish the functional relationship between the allowable holding time of each unit and its node voltage. An event-driven mechanism is used to simulate the dynamic process of each unit being disconnected from the grid in sequence due to the cumulative low voltage time reaching the tolerance limit, generating a step-like response curve of the distribution area power changing with time, i.e., the power response curve.

[0104] Subsequently, to address the internal transient processes of the inverter that the power response curve fails to reflect, a detailed electromagnetic transient model—a single-unit grid-connected model—was established based on the high similarity of the fault ride-through control architecture of different types of inverter interface power generation units. The baseline transient response waveforms of three types of unit operation scenarios during main grid faults were extracted and directly mapped to each unit according to the capacity proportional coefficient. The obtained transient correction was superimposed on the power response curve to obtain a dynamic power response curve for the entire process of voltage drop in the distribution area that takes into account the internal transient processes of the inverter. This achieves rapid simulation that balances simulation accuracy and computational efficiency.

[0105] Step 1: Calculation of initial steady-state power based on amplitude power flow iteration and boundary point iteration update.

[0106] A recursive equation for amplitude power flow, ignoring voltage phase angle, is used in conjunction with a low-voltage ride-through power model for renewable energy units at the inverter interface to establish the voltage-power coupling relationship of a radial distribution area. The voltage cutoff point is defined as the unit number where the first node voltage per unit value is greater than or equal to 0.9. Through iterative updates of the outer cutoff point and a dual-layer power-voltage iterative solution, the steady-state distribution of node voltages for each unit after a voltage drop and the initial power response of the distribution area are calculated. This determines the mixed operating state where some units enter low-voltage ride-through mode while others remain in normal operation. The calculation results serve as the initial state input for event-driven sequential grid disconnection simulation. A summary of the specific steps is as follows: Figure 2 As shown.

[0107] 1. Transformer Area System Model.

[0108] Consider a transformer substation containing The new energy generator sets are connected to the low-voltage side of the transformer in the distribution area via radial busbars. The high-voltage side of the transformer is considered an infinite power grid. The new energy generator sets and lines are numbered sequentially from the low-voltage side outlet of the transformer along the power flow direction:

[0109] (1) No. The grid connection point of the generator unit is considered as a node. Determine the node node voltage and its amplitude per unit value The formula is as follows:

[0110] ;

[0111] In the formula, For node sequence number, The nominal line voltage; where, node For the first The grid connection point for the generator unit.

[0112] (2) Determine the low-voltage side of the transformer in the distribution area as node 0, and determine the node voltage of node 0. per unit value During rated operation .

[0113] (3) Determine the nodes With nodes The line impedance between (busbars) The formula is as follows:

[0114] ;

[0115] In the formula, For nodes With nodes The line resistance between them For nodes With nodes The line reactance between them.

[0116] (4) Calculate the first Rated current of the Taiwanese unit The formula is as follows:

[0117] ;

[0118] In the formula, For the first Rated capacity of the unit.

[0119] (5) During normal operation of each unit ( Constant power output, rated output power is And the power factor angle of each unit Same. Determine the above-mentioned... Rated active power output of the Taiwanese generator set and rated reactive power output The formula is as follows:

[0120] ;

[0121] ;

[0122] Among them, rated reactive power output If positive, it generates inductive reactive power. (The first...) The effective value of the line voltage at the grid connection point of the generator unit is Determine the corresponding reactive current. The formula is as follows: .

[0123] 2. Output power model of new energy units during fault ride-through.

[0124] When the node voltage drops to At this time, the unit enters the Low Voltage Ride Through (LVRT) state; the voltage rises to At this time, the unit enters the High Voltage Ride Through (HVRT) state. During fault ride-through, the grid-connected inverter of the new energy unit outputs dynamic reactive current according to grid connection standards to support the power grid.

[0125] (1) Low voltage ride-through state ( ):

[0126] Calculate the low voltage ride-through period Reference value of the output reactive current of the inverter of the Taiwanese unit The formula is as follows:

[0127] ;

[0128] In the formula, This is the dynamic reactive current proportionality coefficient, typically taken as 1.5 to 3. For the first Rated current of the unit.

[0129] Determine the first Maximum allowable current of inverter in Taiwanese unit The formula is as follows:

[0130] ;

[0131] During fault ride-through, reactive current output is prioritized, while active current is maintained as close to pre-fault levels as possible within the current margin. The inverter's output active and reactive currents during low-voltage ride-through are determined using the following formulas:

[0132] ;

[0133] ;

[0134] ;

[0135] In the formula, For the first The output active current of the inverter in the Taiwanese unit. For the first The output reactive current of the inverter in the Taiwanese unit For the first The active current output of the inverter in the unit before the inverter malfunctions.

[0136] During the fault traversal, the first The output active power of the unit and output reactive power The formula is as follows:

[0137] ;

[0138] ;

[0139] (2) Normal operating status ( ):

[0140] If the node voltage (node ​​voltage) is within the normal range, then the first The generator set is in normal operation, and the inverter output reactive power is... Rated reactive power output Output active power Rated active power output .

[0141] (3) High voltage ride-through state ( ):

[0142] Calculate the first Reference value of output reactive current of inverter in Taiwanese unit The formula is as follows:

[0143] ;

[0144] In the formula, This is the dynamic reactive current proportionality coefficient, typically taken as 1.5 to 3.

[0145] Calculate the first The output reactive current of the inverter in the Taiwanese unit and output active current The formula is as follows:

[0146] ;

[0147] Calculate the first The output reactive power of the inverter in the Taiwanese unit and output active power The formula is as follows:

[0148] ;

[0149] .

[0150] 3. Power flow recursion in radial networks and calculation of total power in transformer areas.

[0151] From the transformer to the end of the busbar Looking at the Taiwanese crew, the statistics show that... The Taiwanese unit and all its downstream units are directed to the node. The formula for the total output power in all directions is as follows:

[0152] ;

[0153] ;

[0154] In the formula, For the first The total active power output of the generator unit and its downstream units. For the first The total reactive power output of the generator unit and its downstream units. For unit serial number, .

[0155] Considering the short busbars between new energy generating units within the same power zone, and that the units generally operate at the same power factor, to avoid phase angle iteration in traditional power flow calculations, reduce computational load while ensuring accuracy, and ignore the influence of the transverse component of voltage drop, the voltage amplitude is calculated based on the longitudinal component of power line voltage drop. The relationship between the voltage magnitudes of adjacent nodes is then calculated using the following formula:

[0156] ;

[0157] In the formula, For nodes node voltage, For nodes With nodes The line resistance between them.

[0158] Taking the low-voltage side of the transformer as the starting point for calculation, the initial amplitude of the voltage at each node is set to the node voltage on the low-voltage side of the transformer in the distribution area. The downstream voltage can be calculated based on the upstream voltage and the power flowing through the line.

[0159] Furthermore, the line current can be approximated to solve for losses. The line impedance can then be calculated. RMS value of the current flowing through it The formula is as follows:

[0160] ;

[0161] Determine the node With nodes The formula for the three-phase power loss of the line is as follows:

[0162] ;

[0163] ;

[0164] Ultimately, the total power received from the low-voltage side of the transformer into the distribution area equals the output power of all generating units minus line losses, as shown in the following formula:

[0165] ;

[0166] ;

[0167] In the formula, This refers to the total active power at the transformer station in the distribution area. This represents the total reactive power at the transformer station in the distribution area.

[0168] Therefore, given the output power of each unit and the line parameters between each node, the total power of the distribution area can be obtained through chain recursion, and the whole process only involves algebraic calculations of the effective values ​​of node voltages.

[0169] 4. Definition of voltage boundary point.

[0170] The power flow direction is from the renewable energy units to the transformer in the distribution area. Due to the line impedance, the voltage rises along the line. During normal operation, the voltage at the node of the unit closest to the transformer is the lowest, and the voltage at the node of the terminal unit m is the highest. A voltage drop occurs on the grid side ( After this, a mixed state may occur where the near-end units of the transformer enter LVRT (Low Voltage Ride Through) while the far-end units continue to operate normally. Define the voltage cutoff point. Unit serial numbers that meet the following conditions:

[0171] ;

[0172] According to the definition, numbering The unit entered the LVRT, numbered The generating units are in normal operating condition. If the node voltage at all generating unit grid connection points... ,but This indicates that all generating units have entered LVRT; if the node voltage at the grid connection point of all generating units... ,but This indicates that all units are operating normally.

[0173] 5. Two-level iterative solution method.

[0174] This invention addresses the power distribution problem of multiple renewable energy generating units in a distribution area after a voltage drop. Since the operating mode of each unit depends on its node voltage, and the node voltage is influenced by the injected power from each unit and the line power flow, these two factors form a coupled nonlinear relationship. This invention employs a two-layer iterative method—outer layer searching for boundary points and inner layer iterative power flow—to solve this problem. The specific steps are as follows:

[0175] Step 1: Set the initial dividing point This assumes that all units are in normal operation after a voltage drop.

[0176] Step 2: Outer layer iteration (the first step) (Time). Based on the current boundary point, allocate the operating status of each unit: if... According to the instructions The Taiwanese unit is in a low-voltage ride-through state; if According to the instructions The unit is in normal operating condition.

[0177] Step 3: Inner layer power flow iteration (the first step) (times), perform the following sub-steps:

[0178] (1) At that time, set the initial voltage value so that the difference in initial node voltage between each node is equal to the node voltage on the low-voltage side of the transformer. The formula is as follows:

[0179] ;

[0180] In the formula, The initial node voltage is denoted as .

[0181] (2) in the In each iteration, the output power of each unit is calculated based on the current voltage amplitude of each node and the operating status of the corresponding unit, including the output reactive power and output active power.

[0182] (3) Calculate the total output power downstream of each branch by accumulating power from the last stage forward, using the following formula:

[0183] ;

[0184] .

[0185] (4) From node To node The updated node voltage magnitudes are calculated based on the total output reactive power and total output active power, using the following formula:

[0186] ;

[0187] In the formula, The node voltage is denoted by . For nodes With nodes The line resistance between them For nodes With nodes The line reactance between them. The updated upstream voltage is directly used, forming a Gauss-Seidel iteration, which helps to accelerate the convergence speed.

[0188] (5) Determine convergence: If , If the value is a minimum or a preset value, then the inner layer converges. Record the voltage of each node and its per-unit value at this time, and continue to step 4; otherwise, let... And return to sub-step (2).

[0189] Step 4: Boundary point update.

[0190] Per-unit values ​​of node voltages obtained from inner-layer convergence Update the dividing point: find the first satisfied point. The unit serial number, if it exists, is recorded as... If the per-unit value corresponding to all nodes Then let This can be represented as:

[0191] ;

[0192] like If the boundary point no longer moves and the outer layer converges, continue with step 5; otherwise, let Return to step 2 and continue the outer iteration.

[0193] Step 5: Result Output. After outer-layer convergence, the node voltage amplitude at all grid connection points of all units. Output active power and output active power Confirmed. Calculate the total output power at the transformer substation using the method in sub-step (3), as follows:

[0194] ;

[0195] ;

[0196] Thus, the voltage amplitude, power distribution, and steady-state total power of each unit at the grid connection point after the voltage drop are obtained, and the output power model of the new energy distribution area during the fault ride-through is obtained, which serves as the initial state for the second step of sequential grid disconnection dynamic simulation.

[0197] Step 2: As Figure 3As shown, the event-driven sequential grid disconnection dynamic simulation of fault ride-through time limits, based on the initial steady state obtained in the first step, introduces the fault ride-through time constraint specified in the grid connection standard to establish a functional relationship between the allowable grid connection time of each unit and its node voltage. The next disconnection event is determined by the unit with the shortest remaining allowable time among all units currently in the low-voltage ride-through state. After executing the disconnection operation, the power flow distribution is recalculated, and the system state is updated after a cascading disconnection judgment. This process is repeated until no new disconnection event occurs, generating a stepped response curve of the transformer area power changing over time. The result of this step serves as the steady-state reference input for the transient correction in the third step.

[0198] 6. Modeling of voltage ride-through time characteristics.

[0199] According to the grid connection standards for new energy generator sets, when the per-unit value of the node voltage... When the voltage drops below 0.9, the unit should be able to operate continuously without disconnecting from the grid for a certain period, the duration of which is determined by the depth of the voltage drop. For the first [unit / stage] in the low voltage ride-through range... Taiwanese generator unit, its continuous operation time without disconnecting from the grid It is a function of its node voltage, as shown in the following formula:

[0200] ;

[0201] In the formula, Per-unit value of node voltage The mapping function to the allowable time is determined by the voltage-time withstand curve specified in the grid connection standard. When the fault duration accumulates to the continuous non-disconnection operating time... At that time, the first The generator unit was disconnected from the power grid and will not be reconnected to the grid for a short period of time.

[0202] 7. The overall approach to event-driven sequential offlineing.

[0203] Assume the node voltage on the low-voltage side of the transformer in the distribution area is The amplitude at time is The voltage drop is assumed to remain constant during the duration of the drop. Using the method in the first step, the steady-state distribution values ​​of the node voltages and unit power before the first unit is switched off are obtained, including the node... Per-unit value of initial node voltage Initial input useful power and initial output reactive power .

[0204] Starting from this steady state The event-driven mechanism is used to simulate the dynamic switching of the time domain:

[0205] For all units in the low voltage ride-through state, from Accumulate fault operation time. When the new energy distribution area reaches the minimum value of the continuous non-disconnection operation time among all nodes, the corresponding unit is disconnected; After the generator unit is disconnected, update the node voltage and non-disconnection continuous operation time of the remaining nodes; repeat steps S22 to S23 until no generator unit is in low voltage ride-through state or the fault is cleared.

[0206] (1) For all units in LVRT status, from Start accumulating fault runtime.

[0207] (2) Each unit calculates its initial non-disconnection continuous operation time based on its corresponding node voltage. Having the smallest The units will be the first in Cut out at any time.

[0208] (3) After the unit is disconnected, the system topology changes, and power flow calculation needs to be performed again to obtain the new node voltage and continuous non-disconnection operation time of the remaining unit.

[0209] (4) At this time, the cumulative fault operation time of the remaining units has reached the value at the time of the cut-off. The per-unit value of the node voltage of each unit The new continuous running time without disconnecting from the network is calculated. If the cumulative time has reached or exceeded If the corresponding unit is shut down, it will be immediately disconnected; otherwise, it will continue to operate and... As the time of the next excision event.

[0210] (5) Repeat this process until no unit is in a low voltage ride-through state or the fault is cleared.

[0211] The entire process will generate a series of discrete power cut-off events. The power of the distribution area remains constant between adjacent events, but changes at the event time, forming a power step response characteristic.

[0212] 8. Iterative calculation steps for dynamic unit disconnection.

[0213] Step 1: Initial Drop Steady-State Calculation. Using the amplitude power flow iterative method considering the boundary point search from Step 1, the node voltage on the low-voltage side of the transformer in the distribution area before the cut-off event is solved. The amplitude of the initial node voltage corresponding to each unit Initial output useful power Initial output reactive power Total active power of the transformer substation Total reactive power The operating status of each unit is determined based on the initial boundary point.

[0214] Step 2: Initialize the timing and event queue. Set the current time when the voltage drop event starts. Calculate the initial non-disconnection continuous operating time for each unit: If the initial node voltage per unit value... ,but Record number The unit has accumulated downtime. If the per-unit value of the initial node voltage ,but No time limit.

[0215] Step 3: Determine the next resection event. Calculate the number of events in the LVRT state. Remaining non-disconnection operating time for the Taiwanese unit The formula is as follows:

[0216] ;

[0217] In the formula, The first refers to the general first The current continuous operating time of the unit without disconnecting from the grid.

[0218] Find the minimum remaining non-disconnected runtime. and the corresponding set of units (Multiple machines may be cut off simultaneously). Therefore, the time of the next cut-off event is calculated based on the current time. The formula is as follows:

[0219] .

[0220] If the units are assembled If the value is empty (no unit is in LVRT and has not switched off), the subsequent state will maintain the current power unchanged until the fault ends, then the calculation will terminate and the iteration will exit.

[0221] Step 4: Advance time and perform the cut. Let the current time be... , will set The units in the process are marked as switched out and will no longer participate in subsequent power flow calculations. Meanwhile, all units still in operation and in LVRT status... The cumulative downtime of the generator set has been updated to: .

[0222] Step 5: New Steady-State Calculation After Removal. Under the new topology after removing the disconnected units, re-execute the amplitude power flow iteration calculation from Step 1 to obtain the updated boundary points and the amplitude of the node voltages of each operating unit. and output active power and output reactive power In power flow calculations, the power model (LVRT / normal operation state) is still dynamically selected based on the node voltage of each unit. At this time, the newly entered LVRT... Cumulative downtime of the unit .

[0223] Step 6: Chain Removal Condition Check. After entering the new steady state, for each still running [machine name], [the following steps are performed]. Taiwanese generator set, if its corresponding per-unit value Then recalculate the corresponding new non-disconnection running time. and cumulative downtime Compare.

[0224] like This indicates the first The unit has met the disconnection criteria after voltage redistribution and should be disconnected immediately, joining the interlocked disconnection group. .

[0225] If the chain resection set If the value is not empty, immediately disconnect the corresponding unit and update the topology, then return to step 5 to recalculate the new steady state, until no new chain disconnection occurs, i.e., the chain disconnection set is reached. Empty.

[0226] If the interlocking resection set Empty, update the continuous non-disconnection operating time of each unit. Proceed to step 7.

[0227] Step 7: Record the power and continue. At the current moment... The total active power value under the new steady state is obtained by updating. Total active power value This value will remain until the next excision event. Return to step 3 and continue searching for the next excision event.

[0228] 9. Generation of power response curves for transformer substations.

[0229] Through the above iterative steps, a series of discrete cut-off event times 0 < t1 < t2 < ... and the corresponding steady-state power values ​​of the transformer area can be obtained:

[0230] ;

[0231] In the formula, This is the steady-state active power curve of the new energy distribution area. The steady-state reactive power curve of the new energy distribution area. This is the first time the steady-state active power of the new energy distribution area has been cut off. This is the first time the steady-state reactive power of the previous renewable energy distribution area has been cut off. This represents the steady-state active power of the renewable energy distribution area after the first disconnection. This represents the steady-state reactive power of the renewable energy distribution area after the first disconnection. The time of the first excision. This refers to the time of the second excision.

[0232] Total active power value The power remains constant between adjacent events, but changes abruptly at the moment of the event, forming a complete stepped power response curve. This curve reflects the dynamic power change process in a distribution area with multiple new energy units during a sustained voltage drop, caused by the sequential disconnection of each unit from the grid based on its own node voltage and fault crossover time limit. It can be used as an equivalent power injection model for transient stability analysis of the upper-level power grid.

[0233] The above method transforms the complex dynamic process that originally required time-domain integration into a discrete event sequence consisting of a finite number of steady-state power flow calculations and simple algebraic judgments. This avoids the high computational cost of small-step numerical integration, significantly improves computational efficiency, provides clear physical meaning, and facilitates engineering implementation.

[0234] Step 3: As Figure 4 As shown, inverter transient response correction is based on detailed single-unit mechanism simulation. Building upon the power stepped response curves obtained in the first two steps, and considering the millisecond to tens of millisecond transient responses generated by the inverter's internal control loop during voltage surges, a detailed electromagnetic transient simulation model is established for only one unit, leveraging the high consistency of control parameters and main circuit structure between inverters in the same renewable energy unit area. Two types of benchmark transient response waveforms are extracted and directly mapped to each unit using a capacity scaling factor. The transient correction is then superimposed onto the stepped response curve, achieving accurate characterization of the inverter's internal transient processes with minimal additional computational cost.

[0235] 10. Power response curve of the transformer area considering transient characteristics.

[0236] The first and second steps respectively obtained the initial steady-state power distribution after the voltage drop and the dynamic power step curve of the unit sequential cutoff considering the fault ride-through time. However, both of these methods treat the inverter's power mode switching as instantaneous, neglecting the transient adjustment process caused by the inverter's internal current loop PI controller and output filter. Existing research shows that the power overshoot or damped oscillations caused by this transient process are not negligible in the short-timescale system dynamic analysis after a fault, affecting the action criteria of protection devices and stability margin assessment. If a detailed electromagnetic transient model is established for simulation of each unit, the computational scale will be too large, losing the advantage of rapid simulation.

[0237] Considering that inverters for new energy generating units within the same distribution area are typically supplied by the same manufacturer, their control parameters and main circuit structures are highly consistent. Furthermore, the transient response characteristics of the same inverter model at different capacity levels are primarily determined by capacity, with the amplitude of the transient response scaling approximately linearly with the unit capacity. Within the same distribution area, although the voltage drop depth at the grid connection point of each unit varies along the line, this difference is generally small within a limited line impedance range, and its impact on the transient response waveform is negligible. Therefore, under the condition that only capacity differences need to be considered between different units, this invention proposes an equivalent transient response correction method based on single-unit mechanism simulation and capacity ratio mapping. The specific implementation steps are as follows:

[0238] Step 1: Establish a detailed mechanism simulation model (single-unit grid-connected model) for a single renewable energy unit. Select the first renewable energy unit in the distribution area (closest to the transformer) as the baseline unit. Build its single-unit grid-connected model in an electromagnetic transient simulation platform (such as Simulink), including a three-phase inverter bridge, grid-connected filter, PI controller, and fault ride-through control logic, etc. Parameters are provided by the power plant or selected from typical values. The capacity of this simulation model is denoted as the baseline capacity. .

[0239] Step 2: Extraction of the three types of baseline transient responses.

[0240] The solution scenario is Voltage drops occur frequently. Voltage recovery at specific moments. Taking a benchmark unit as the object, simulations were performed under three typical voltage surge scenarios, and the corresponding benchmark transient response waveforms were extracted. This benchmark response waveform characterizes the transient transition process of the inverter under the corresponding event type at a unit capacity. The actual transient correction amount can be obtained by scaling each unit according to its own capacity.

[0241] Scenario A: The unit's baseline transient response enters LVRT at the start of the drop. The node voltage of unit 1 after the drop has been obtained in the first step mentioned above. In the stand-alone grid-connected model, the initial operating state is set as follows: the node voltage is the normal operating value before the fault. The inverter operates stably at rated power. The node voltage is kept at... The voltage drops to the node voltage at any time. ,exist Restore to normal operating value at all times Record the time-domain response curves of the inverter's output active and reactive power, denoted as . and Its steady-state value during the fault period is denoted as and This response waveform serves as the reference transient response waveform for all units entering the LVRT state.

[0242] Scenario B: Baseline transient response of a unit that remains in normal operation at the start of the voltage drop. The node voltage threshold is determined in the first step. ,serial number to The unit node voltage still meets The unit is within its normal operating range. Although the grid connection point of this type of unit is within the normal operating range, the amplitude has dropped compared to before the fault, and the internal control loop will still produce a small transient response. Among them, the first... The node voltage before the drop of the first unit (the first unit after the dividing point) is: After falling, In the stand-alone grid-connected model, the initial operating voltage is set to the level before the voltage drop. and rated power condition, in The moment fell to , Time to return to Record the time-domain response curve and The steady-state value of the waveform that serves as the reference transient response is the rated power.

[0243] Scenario C: The unit is in normal operation at the start of the voltage drop, and the baseline transient response during the fault period enters the LVRT (Low Voltage Reduction Zone). This scenario applies to units that newly enter the LVRT due to a voltage drop below 0.9 pu caused by a sequential grid disconnection event. A representative unit of this type, with its capacity or voltage characteristics, is used as the simulation object. The voltage step transition to the LVRT region and the recovery from the LVRT region to normal operation are simulated in a single-unit grid-connected model, and the power response waveforms are recorded. and This serves as the corresponding reference transient response waveform and as the reference for the transient state of this type of unit mode switching.

[0244] Step 3: Capacity proportional coefficient mapping and transient correction superposition. For any i-th renewable energy unit in the distribution area, define a capacity proportional coefficient to map the reference transient response to the actual transient response of the unit.

[0245] Calculate the capacity ratio factor The formula is as follows:

[0246] ;

[0247] This reflects the linear scaling relationship between the capacity difference between the unit and the reference unit and the transient response amplitude. Each unit selects the corresponding reference transient response waveform based on the type of voltage surge event it has experienced, and directly scales it according to the capacity scaling factor to obtain the transient correction amounts for the output active and reactive power of each unit.

[0248] Each unit selects the corresponding reference transient response based on the type of voltage surge event it has experienced, and directly scales it according to the capacity scaling factor to obtain the transient correction amount, as shown in the following formula:

[0249] ;

[0250] ;

[0251] In the formula, For the first The transient correction of the active power output of the generator unit. For the first The transient correction amount of the reactive power output of the generator unit. For the first Capacity ratio coefficient of the unit; The time-domain response curve for output active power includes... , and ; This represents the steady-state active power of the single-unit grid-connected model during voltage dips; The time-domain response curve for output reactive power includes , and ; This represents the steady-state reactive power of the single-unit grid-connected model during voltage dips.

[0252] The physical meaning of the above correction amount is: the dynamic deviation of the inverter output power from the moment of the voltage surge event until the re-establishment of steady state, relative to the final steady-state value. This deviation has been scaled proportionally to the capacity and decays to zero over time; the correction amount automatically returns to zero after steady-state is reached.

[0253] Step 4: Superimpose the transient correction and the stepped power curve to obtain the steady-state active power curve from step 2. and steady-state reactive power curve As a steady-state reference power. At each voltage surge event. (e.g., the start of the fall) Timing of each unit's disconnection Voltage recovery (etc.) to determine the set of units affected by the event. Based on the voltage change type, the transient corrections for each unit are shifted over time and then superimposed onto the steady-state reference power:

[0254] ;

[0255] ;

[0256] In the formula, The dynamic power response curve for the active power output of the new energy distribution area. The dynamic power response curve for reactive power output from the new energy distribution area. For the occurrence time of each voltage surge event, This refers to the set of generating units involved in voltage surge events. When a voltage surge event occurs, the first The transient correction of the active power output of the generator unit. When a voltage surge event occurs, the first The transient correction of reactive power output by the generator unit; among which, voltage surge events include voltage sag, unit disconnection and voltage recovery.

[0257] The obtained dynamic power response curve and This is the complete power dynamic response curve of the transformer area after taking into account the transient process of the inverter's internal control. This curve is consistent with the step curve of the second step during the steady-state time period. It superimposes an accurate transient transition process near the voltage change moment. It has a high degree of accuracy with the full electromagnetic transient simulation on the millisecond time scale, while the computational cost is only equivalent to a few steady-state power flow solutions plus a detailed simulation of a single unit, thus achieving an effective balance between simulation accuracy and computational efficiency.

[0258] This invention determines the initial hybrid steady-state operation after voltage dips, considering the voltage distribution differences within the transformer substation, through amplitude power flow iteration and boundary point search in the first step. Based on this, the second step introduces fault ride-through time constraints, simulating the sequential grid disconnection dynamic process of the units using an event-driven mechanism to generate a power step response curve. Furthermore, the third step utilizes detailed single-unit mechanism simulation to extract the baseline transient response, which is then mapped and superimposed onto the step curve via a capacity proportional coefficient, compensating for the neglect of internal inverter transients in the first two steps. These three steps are progressive and interconnected, forming a complete simulation chain from steady-state distribution to discrete event dynamics and then to continuous transient correction, comprehensively depicting the full-process dynamic power characteristics of high-proportion renewable energy transformer substations under voltage dip scenarios.

[0259] Example 2:

[0260] Taking a radial distribution area with a high proportion of photovoltaic (PV) units as an example, after being equivalent to a single busbar, the units are numbered from closest to furthest from the transformer in the distribution area. The capacity of each unit and the impedance between units are shown in Table 1. This distribution area contains 12 PV units with a total PV capacity of 14MW. Residential static loads are connected to the grid connection points of units numbered 3, 5, 8, 10, and 12. Each unit has a power of 300~400kW and a power factor of 0.88. Using the ZIP static load polynomial model, the total load is 1.70 MW + j0.92MVAR.

[0261] Table 1. Distribution of Photovoltaic Unit Capacity and Distance

[0262]

[0263] The fault scenario is set as follows: the power grid on the high-voltage side of the transformer is... Voltage drops occur frequently; grid voltage per unit value The price plummeted from 1.0 PU before the drop to 0.83 PU, and the drop continued until... The voltage then recovered to 1.0 pu, and the simulation terminated. Dynamic reactive current proportional coefficient for each unit Inverter maximum allowable current Take 1.05 pu. The low-voltage ride-through time for each unit is as follows: Figure 5 The voltage envelope shown is determined.

[0264] The steady-state distributions of the grid-connected voltages at each node before and after the voltage drop, calculated using the amplitude power flow iteration method based on boundary point search in the first step of Example 1, are as follows: Figure 6 , Figure 7 As shown in the figure. The horizontal axis represents the unit number ( The vertical axis represents the per-unit voltage at the grid connection point. The hollow dot curve represents the unit voltage distribution before the voltage drop, while the solid dot curve represents the unit voltage distribution after the voltage drop.

[0265] Depend on Figure 2 It can be seen that:

[0266] (1) Before the fault, the voltage at the grid connection point of each unit gradually increased from the near end to the far end, with the voltage of the terminal unit being the highest.

[0267] (2) After the fault, the node voltage on the low-voltage side of the transformer suddenly dropped to 0.83 pu, and the grid connection point voltage of the near-end units (units 1 to 5) dropped below 0.9 pu, entering the LVRT state; the grid connection point voltage of the far-end units (units 6 to 12) remained above 0.9 pu, and were in normal operation. This mixed operating state verified the basic judgment in Example 1 regarding the differential response of units caused by the difference in voltage distribution within the transformer area. Figure 7 By dividing point The vertical axis serves as the dividing line, visually displaying the search results by division. Units 1 to 5 on the left side of the dividing line are operating in LVRT mode, while units 6 to 12 on the right side are operating in normal mode.

[0268] The voltage distribution curves and the number of iterations for inner and outer layers during the iteration process are as follows: Figure 8 As shown. By Figure 8 It can be seen that the outer boundary point converges to the boundary point after 3 iterations. The inner power flow iteration requires approximately 4-8 iterations per step, totaling 16 iterations, with a convergence threshold of [missing information]. pu indicates that the amplitude power flow two-layer iterative method has good convergence characteristics.

[0269] The distribution of active and reactive power output of each unit before and after the voltage drop is as follows: Figure 9 As shown in the figure, blank bars represent the power output of each unit before the fault, and striped bars represent the steady-state power output after the fault. It can be seen that units 1-5, which are in LVRT (Low Voltage Reduction) state, have significantly reduced active power output compared to before the fault (constrained by inverter current limiting and priority reactive power support strategy), while simultaneously injecting reactive power into the grid to support the grid voltage. The reactive power output of each unit varies depending on the voltage drop depth at the grid connection point and the unit capacity. Units 6-12, which are in normal operation, maintain their rated active power output, while their reactive power output is zero. This result verifies that the method in the first step can accurately solve the steady-state power distribution of each unit under mixed operation considering the differences in voltage distribution within the distribution area.

[0270] The step response curves of the total active power and total reactive power of the transformer area as a function of time generated by the event-driven method in the second step of Example 1 are shown below. Figure 10 As shown. To clearly display the dynamic details of transient events, a long period of constant power is compressed using a broken-shaft method, divided into three display windows: the initial voltage drop segment (0~0.4s), the sequential grid disconnection segment (1.55~1.7s), and the voltage recovery segment (1.95~2.1s). The top row of images shows active power, and the bottom row shows reactive power. Black dashed lines and letters indicate the timing of each voltage surge event. It can be seen that: in After a voltage drop, the total power of the distribution area immediately jumps from the rated operating state before the fault to a mixed operating state. Due to the relatively small voltage drop, reactive power mainly depends on the reactive power consumed by line impedance. As each unit switches off in sequence according to the fault ride-through time limit, the active power of the distribution area decreases step by step, and the net reactive power absorption also decreases accordingly. The power remains constant between each switching event, and a step change occurs at the moment of switching off, forming a clear step-like response characteristic. The switching sequence is unit 1→2→3→4, with the time interval decreasing successively, reflecting the accelerating effect of cascading switching. When the voltage recovers, only 8 units remain in normal operation. This stepped curve can be directly used as the equivalent power injection model for transient stability analysis of the upstream power grid.

[0271] The three types of baseline transient response waveforms (per-unit values ​​based on rated capacity) extracted in the third step of Example 1 based on the detailed single-machine model simulation are as follows: Figure 11As shown, the horizontal axis represents time, with the top row representing active power response and the bottom row representing reactive power response. Type 1 (Scenario A) is the transient response of the unit entering LVRT at the start of a voltage dip. Type 2 (Scenario B) is the transient response of the unit maintaining normal operation at the start of a voltage dip, with an overall amplitude significantly greater than Type 1 (Scenario A) and Type 3 (Scenario C). Type 3 (Scenario C) is the transient response of the unit entering LVRT from normal operation during a fault, with a response shape similar to Type 1 (Scenario A) but a smaller amplitude. The differences among the three types of transient responses verify the necessity of constructing reference responses according to event type to characterize the inverter control dynamics in different scenarios. The above three sets of waveforms are the reference transient corrections for subsequent capacity proportional mapping.

[0272] Will Figure 11 The transient correction amount is equivalent to the capacity. Figure 10 After analyzing the steady-state power response curves of each unit, the following results are obtained: Figure 12 The power response curves after transient correction for the transformer area are shown. It can be seen that near each voltage surge event (voltage drop, disconnection, recovery), the corrected power curve exhibits a smooth transition, overcoming the instantaneous jump defect of the stepped curve at the event moment. The transient correction only works within a short time window (approximately 20-30 ms) after the event, then decays to zero. During the steady-state period, the two curves completely overlap, indicating that the correction method only supplements the neglected internal dynamics of the inverter at the required time scale, without affecting the accuracy of steady-state power calculation. Figure 13 As shown in the full time axis perspective, the overall evolution trend of the power in the distribution area is clear: steady state before the drop → mixed operation state of the units → step response of the unit interlocking switch → steady state after recovery.

[0273] Analysis of this embodiment shows that the three-step progressive rapid simulation method proposed in this invention can accurately characterize the dynamic power characteristics of a high proportion of renewable energy distribution areas throughout the entire process under voltage drop scenarios:

[0274] The first step effectively solved the initial mixed steady state of operation, taking into account the differences in voltage distribution within the transformer area and the static load voltage dependence characteristics. In the case study, it only required 3 outer iterations and 16 inner iterations to converge to the boundary point.

[0275] The second step simulates the dynamic process of sequential grid disconnection of units under fault travel time constraints using an event-driven approach. It accurately captures the event timing of step-by-step disconnection and operation state switching of each unit, generating power ladder response curves with clear physical meaning.

[0276] The third step is a capacity ratio mapping method based on a detailed single-unit mechanism model. With minimal additional computational cost (only simulating 3 sets of baseline transient responses and superposition calculations are required), the transient processes inside the inverter are incorporated into the simulation model, accurately reflecting the transient responses generated by the switching of unit operating states.

[0277] The entire method improves computational efficiency while maintaining simulation accuracy: steady-state solutions require only about 16 amplitude power flow recursions, event-driven simulations require only 4 event steps of steady-state recalculation, and transient corrections only require capacity scaling and algebraic superposition of 3 types of reference responses. Compared to all-electromagnetic transient simulations (which require...), this method significantly improves computational efficiency. By establishing detailed electromagnetic models of each inverter unit and simultaneously solving differential-algebraic equations, an acceleration of approximately two orders of magnitude can be achieved, meeting the timeliness requirements for rapid simulation of a large number of scenarios in system-level fault scanning, relay protection setting, and stability control strategy formulation.

[0278] In summary, compared to existing technologies, the technical solution provided by this invention effectively represents the initial mixed-operation steady state considering the voltage distribution differences within the renewable energy distribution area and the dynamic load voltage dependence characteristics by constructing an output power model. It accurately depicts the event sequence of step-by-step disconnection and operating state switching of each unit by constructing power response curves, and incorporates the transient processes within the inverter into the simulation model through a single-unit grid-connected model, accurately reflecting the transient responses generated by unit operating state switching. In short, the technical solution provided by this invention can accurately characterize the full-process dynamic power characteristics of a high-proportion renewable energy distribution area under voltage drop scenarios.

Claims

1. A simulation method for the dynamic characteristics of new energy distribution areas under main grid faults, characterized in that, Includes the following steps: Step S1: Set the low-voltage side node of the transformer in the distribution area as the initial boundary point of the node voltage and update iteratively to construct the output power model of the new energy distribution area during fault ride-through. Step S2: Determine the disconnection sequence based on the continuous non-disconnection operation time of each unit; based on the output power model, calculate the output reactive power and output active power of the new energy distribution area at each stage according to the disconnection sequence of each unit, and construct the power response curve; Step S3: Select the unit closest to the transformer in the new energy distribution area as the benchmark unit, and build a single-unit grid-connected model on the electromagnetic transient simulation platform; Step S4: Using the benchmark unit as the object, perform simulations of various voltage change scenarios and extract the corresponding benchmark transient response waveforms; Step S5: Based on the reference transient response waveform, calculate the transient correction amount of output active power and output reactive power in each unit according to the ratio of the reference capacity to the rated capacity of the single-unit grid-connected model. Step S6: Superimpose the transient correction onto the power response curve to obtain the dynamic power response curve of the new energy distribution area.

2. The simulation method for the dynamic characteristics of new energy distribution areas under main grid faults according to claim 1, characterized in that, Step S1 includes: Step S11: Set all units to be in normal operation after a voltage drop occurs, and use the low-voltage side of the transformer in the distribution area as the initial dividing point. ; Step S12: Includes steps S121 to S124; Step S121: Set the initial amplitude of each node voltage to the node voltage on the low-voltage side of the transformer in the distribution area. ; Step S122: Calculate the output reactive power and output active power of the corresponding unit based on the voltage amplitude of each node and the operating status of the corresponding unit; Step S123: Calculate the total output reactive power and total output active power of each downstream unit at each node; Step S124: Update the voltage of each node based on the total output reactive power and the total output active power, and calculate the difference in the voltage amplitude of the corresponding node before and after the update; If all differences are less than the preset value, continue; otherwise, return to step S12. Step S13: Update the boundary points according to the per-unit values ​​of the updated node voltages; If the updated boundary point is equal to the original boundary point, continue; otherwise, return to step S12. Step S14: Based on the output reactive power and output active power of each unit, calculate the total reactive power and total active power at the transformer in the distribution area to obtain the output power model.

3. The simulation method for the dynamic characteristics of new energy distribution areas under main grid faults according to claim 2, characterized in that, Step S122 includes: Determine the operating status of each unit: if node Less than the current dividing point Then the first The unit is in a low-voltage ride-through state; otherwise, the first The unit is in normal operating condition; Calculate the first The output reactive power and output active power of each generating unit include: (1) If the first The generator unit is in a low voltage ride-through state; calculate the output reactive power. and output active power The formula is as follows: ； ； ； ； ； ； ； In the formula, For node sequence number, , For the number of nodes, For nodes The node voltage amplitude, For the first The output active current of the inverter in the Taiwanese unit. For the first The output reactive current of the inverter in the Taiwanese unit. For the first Reference value of the output reactive current of the inverter in the Taiwanese unit. For the first The maximum allowable current of the inverter in the Taiwanese unit. For the first The output active current of the inverter in the Taiwanese unit before the inverter failure. For the first The initial current of the generator set This is the dynamic reactive current proportionality coefficient. For nodes The per-unit value of the node voltage. For the first The rated current of the unit For the first Rated active power output of the generator set For the first The node voltage amplitude before the inverter failure in the Taiwanese unit. This is the nominal line voltage; (2) If the first If the generator set is in normal operating condition, then the first... The output reactive power of the inverter in the Taiwanese unit For the first Rated reactive power output of the unit Output active power For the first Rated active power output of the Taiwanese generator set ; The formula for step S123 is as follows: ； ； In the formula, For the first The total active power output of the generator unit and its downstream units. For the first The total reactive power output of the generator unit and its downstream units. For unit serial number, ; The formula for updating the voltage of each node based on the total output reactive power and the total output active power is as follows: ； In the formula, The node voltage is denoted as . For nodes With nodes The line resistance between them For nodes With nodes Line reactance between; The step of updating the boundary point based on the per-unit value of each node voltage includes: per unit value The first node serves as the new dividing point. If the per-unit value corresponding to all nodes Then let the dividing point ; The formula for step S14 is as follows: ； ； ； ； ； In the formula, This refers to the total active power at the transformer station in the distribution area. This represents the total reactive power at the transformer substation. For nodes With nodes The three-phase active power loss of the line between them. For nodes With nodes The three-phase active power loss of the line between them. Line impedance The effective value of the current flowing through it.

4. The simulation method for the dynamic characteristics of a new energy distribution area under a main grid fault according to claim 3, characterized in that, Step S2 includes: Step S21: Set the voltage drop on the low-voltage side of the transformer in the distribution area when... Based on the node voltage of the low-voltage side of the transformer in the distribution area Calculate the node voltage, output reactive power and output active power of each grid connection point before the first unit is disconnected according to step S122. For all units in the low voltage ride-through state, from Accumulate downtime during fault operation; Step S22: When the new energy distribution area reaches the minimum value of the continuous non-disconnection operation time of each node, the corresponding unit is disconnected; Step S23: After the generator unit is disconnected, update the node voltage and non-disconnection continuous operation time of the remaining nodes; Step S24: Repeat steps S22 to S23 until no unit is in a low voltage ride-through state or the fault is cleared; The power response curve is constructed using the following formula: ； In the formula, This is the steady-state active power curve of the new energy distribution area. The steady-state reactive power curve of the new energy distribution area. This is the first time the steady-state active power of the new energy distribution area has been cut off. This is the first time the steady-state reactive power of the previous renewable energy distribution area has been cut off. This represents the steady-state active power of the renewable energy distribution area after the first disconnection. This represents the steady-state reactive power of the renewable energy distribution area after the first disconnection. The time of the first excision. This refers to the time of the second excision.

5. The simulation method for the dynamic characteristics of new energy distribution areas under main grid faults according to claim 4, characterized in that, It also includes a method for updating the continuous online running time, the formula of which is as follows: ； In the formula, For nodes The continuous operating time without disconnecting from the network. Per-unit value of node voltage A mapping function to the allowed time.

6. The simulation method for the dynamic characteristics of a new energy distribution area under a main grid fault, as described in claim 4, is characterized in that... Step S4 includes: Scenario A: For units that enter the low-voltage ride-through state synchronously with a voltage dip: In the single-unit grid-connected model, the grid connection point voltage is set to the normal operating value of node 1 before the fault, and the inverter operates stably at rated power; Let the grid connection point voltage be at When the voltage drops to the node voltage of node 1 after the voltage drop of the first generating unit, When the inverter returns to the normal operating value of node 1, the time-domain response curves of the output active power and output reactive power of the inverter are recorded as the corresponding reference transient response waveforms. Scenario B: For units that are still operating normally when the voltage drops: Update the boundary point according to step S13; In the stand-alone grid-connected model, the grid connection point voltage is set to the normal operating value at the fault boundary point, and the inverter operates stably at its rated power. Let the grid connection point voltage be at When the voltage drops to the node voltage at the breakpoint after the corresponding unit experiences a voltage drop, When the inverter returns to its normal operating value at the boundary point, record the time-domain response curves of the inverter's output active power and output reactive power as the corresponding reference transient response waveforms. Scenario C: For units that are still operating normally when the voltage drops, but enter a low-voltage ride-through state during the voltage drop: In the stand-alone grid-connected model, the simulations of scenario B and scenario A are executed sequentially, and the time-domain response curves of the inverter's output active power and output reactive power are recorded as the corresponding reference transient response waveforms.

7. The simulation method for the dynamic characteristics of a new energy distribution area under a main grid fault according to claim 6, characterized in that, The formula for step S5 is as follows: ； ； ； In the formula, For the first The transient correction of the active power output of the generator unit. For the first The transient correction amount of the reactive power output of the generator unit. For the first The capacity ratio factor of the Taiwanese unit, To output the time-domain response curve of active power, This represents the steady-state active power of the stand-alone grid-connected model during voltage dips. The time-domain response curve for output reactive power. This represents the steady-state reactive power of a single-unit grid-connected model during a voltage dip. For the first Rated capacity of the unit This is the base capacity corresponding to the base unit; The formula for step S6 is as follows: ； ； In the formula, The dynamic power response curve for the active power output of the new energy distribution area. The dynamic power response curve for reactive power output from the new energy distribution area. For the occurrence time of each voltage surge event, This refers to the set of generating units involved in voltage surge events. When a voltage surge event occurs, the first The transient correction of the active power output of the generator unit. When a voltage surge event occurs, the first The transient correction of reactive power output by the generator unit; among which, voltage surge events include voltage sag, unit disconnection and voltage recovery.

Images