Network-formation type energy storage fault network-formation control method based on coupling impedance safety domain optimization

Abstract

The embodiment of the application provides a network-constructing energy storage fault network-constructing control method based on coupling impedance security domain optimization, comprising the following steps: establishing a network-constructing energy storage grid-connected system and a voltage drop fault model, collecting three-phase voltages and currents of a PCC and performing positive and negative sequence decomposition, and identifying symmetric voltage drop and asymmetric voltage drop working conditions; then, a positive and negative sequence coupling output impedance model is constructed, and a fault network-constructing stability security domain is constructed; a multi-objective optimization function is established, a stable security domain guided multi-strategy improved snake swarm optimization algorithm is adopted to jointly optimize positive and negative sequence virtual impedance, VSG fault inertia, damping coefficient, reactive power support coefficient and active power derating coefficient, and the optimization parameters are embedded into a VSG control link. The method can effectively suppress the fault current peak value during voltage drop, improve the current imbalance problem under asymmetric fault, and improve the fault network-constructing stability of the network-constructing energy storage.

Description

Technical Field

[0001] This invention belongs to the field of new energy control and smart grid technology, and particularly relates to a fault control method for grid-type energy storage based on coupling impedance security domain optimization. Background Technology

[0002] With the continuous increase in the penetration rate of new energy sources and the proportion of power electronic equipment, power systems are exhibiting characteristics such as low inertia, weak damping, and low short-circuit capacity. Grid-based energy storage converters (such as virtual synchronous generator (VSG) control) can autonomously construct voltage frequencies and simulate the inertia and damping of synchronous motors, becoming a key technology for improving the safety and stability of power grids with a high proportion of new energy sources. However, grid-based energy storage is essentially a low-output impedance voltage source. When the grid experiences symmetrical or asymmetrical voltage drop faults, the sudden increase in voltage difference leads to a rapid rise in fault current, which may exceed the device's tolerance range and affect continuous grid operation.

[0003] Existing fault ride-through methods include current limiting, virtual impedance, internal potential regulation, and grid connection / grid connection mode switching. Current limiting can easily alter the characteristics of the VSG voltage source and weaken its grid connection support capability; mode switching relies on phase-locked loops and external grid references, and is prone to losing grid connection capability under weak grid conditions; virtual impedance can increase the equivalent electrical distance to suppress fault current and has good engineering applicability, but traditional designs are mostly based on empirical tuning of current amplitude or drop depth, mainly focusing on current limiting effect and lacking constraints on the stability of the impedance interaction between the converter and the grid. If the impedance is too small, the current limiting is insufficient; if it is too large, it may cause oscillations and reduce reactive power support or power angle stability margin.

[0004] Further analysis reveals that symmetrical faults lead to a decrease in positive-sequence voltage support and an increase in positive-sequence current; asymmetrical faults also generate negative-sequence voltage / current, causing problems such as three-phase imbalance and second-harmonic power fluctuations. Traditional positive and negative-sequence control processes the positive and negative-sequence channels separately or uses only the negative-sequence virtual impedance, failing to fully consider their coupling relationship—negative-sequence current can negatively impact the positive-sequence support capability through limiting, power calculation, and voltage control. Relying solely on a single impedance margin or current-limiting index is insufficient to characterize the coupling stability boundary under asymmetrical faults, and it also fails to simultaneously address multiple objectives such as overcurrent suppression, reactive power support, power angle stability, and negative-sequence suppression.

[0005] At the parameter optimization level, existing methods mostly rely on human experience, fixed virtual impedance, or traditional intelligent optimization. The objectives are often single indicators such as peak current and frequency / voltage deviation, lacking comprehensive consideration of impedance stability boundaries, positive and negative sequence coupling strength, disturbance amplification risk, and synchronization stability constraints. Conventional optimization algorithms are prone to premature convergence, getting trapped in local optima, or generating infeasible individuals. This results in optimized parameters that can suppress overcurrent under certain types of faults but reduce stability margin or negative sequence suppression capability under other operating conditions. Summary of the Invention

[0006] The fault grid control method for grid-type energy storage based on coupling impedance safety domain optimization provided in this application can effectively suppress the peak fault current during voltage dips, improve the current imbalance problem under asymmetrical faults, and enhance the stability of grid-type energy storage fault grid.

[0007] In a first aspect, embodiments of this application provide a fault grid control method for grid-type energy storage based on coupling impedance security domain optimization, including:

[0008] S1. Establish a grid-connected energy storage system and voltage sag fault model, and determine the mathematical relationship between the output voltage of the energy storage converter, the filter inductor current, the point of common coupling voltage, the grid current, the grid equivalent impedance, the voltage sag depth, the positive and negative sequence disturbance components and the control parameters of the virtual synchronous generator (VSG).

[0009] S2. Collect the three-phase voltage and three-phase current at the point of common coupling, and decompose the three-phase voltage and three-phase current into positive and negative sequence to obtain positive sequence voltage, negative sequence voltage, positive sequence current and negative sequence current; based on the positive sequence voltage drop depth and negative sequence imbalance coefficient, identify whether the current fault is a symmetrical voltage drop condition or an asymmetrical voltage drop condition.

[0010] S3. Based on positive-sequence voltage, negative-sequence voltage, positive-sequence current and negative-sequence current, establish a positive and negative sequence coupling output impedance model for grid-type energy storage during voltage drop faults. Incorporate the inherent output impedance of VSG, positive-sequence virtual impedance, negative-sequence virtual impedance, positive and negative sequence coupling compensation impedance, filter impedance, voltage and current inner loop and digital control delay into a unified impedance modeling framework to form a positive and negative sequence coupling output impedance matrix.

[0011] S4. Based on the positive and negative sequence coupled output impedance model and the identified operating conditions, a fault network stability and safety domain is constructed. The fault network stability and safety domain includes the generalized impedance characteristic root stability distance, impedance closed-loop sensitivity peak value, positive and negative sequence coupling strength, characteristic frequency drift, VSG power angle synchronous stability constraint, reactive power support constraint and fault current boundary constraint.

[0012] S5. Establish a multi-objective optimization function with the goals of fault overcurrent suppression, impedance stability improvement, positive and negative sequence coupling reduction, reactive power support enhancement, power angle offset suppression, and parameter coordination improvement.

[0013] S6. Improve the snake swarm optimization algorithm using a stable safety domain-guided multi-strategy approach. With a multi-objective optimization function as the optimization objective and the stable safety domain of fault-based network as the constraint, perform joint parameter optimization on positive-sequence virtual impedance, negative-sequence virtual impedance, positive and negative-sequence coupling compensation impedance, virtual inertia during fault, fault damping coefficient, reactive power support coefficient, and active power derating coefficient to obtain the optimal combination of fault-based network control parameters.

[0014] S7. The optimal fault grid-connection control parameter combination is embedded into the active-frequency control loop, reactive-voltage control loop and positive-negative sequence virtual impedance control loop of the virtual synchronous generator, and combined with the preset fixed current safety upper limit applied to the current vector limiting loop, to generate the three-phase voltage reference value of the grid-connected energy storage converter during the voltage drop fault, so as to perform fault grid-connection control.

[0015] In one optional implementation, S1 specifically includes:

[0016] Establish a DC-side model for energy storage, an AC-side filter model for grid-type energy storage, and a grid equivalent impedance model;

[0017] Establish symmetrical voltage dip fault models and asymmetrical voltage dip fault models;

[0018] A VSG control model is established, which includes active-frequency control equations, reactive-voltage control equations, active power derating control equations during faults, and reactive power voltage support equations during faults.

[0019] In an optional implementation, in S2, identifying the fault type based on the positive-sequence voltage drop depth and the negative-sequence imbalance coefficient includes:

[0020] Calculate the positive sequence voltage drop depth:

[0021] ;

[0022] in, This represents the positive sequence voltage drop depth. The positive sequence voltage magnitude at the point of common coupling; This is the rated voltage amplitude;

[0023] Calculate the negative order imbalance coefficient:

[0024] ;

[0025] in, It is a negative order imbalance coefficient. The negative sequence voltage amplitude at the point of common coupling; To prevent small positive numbers with a denominator of zero;

[0026] like and If so, it is determined to be a symmetrical voltage drop condition;

[0027] like and If the voltage drops, it is determined to be an unbalanced voltage drop condition; among which, A threshold for identifying voltage dips; Threshold for identifying negative order imbalance.

[0028] In one optional implementation, the step of forming the positive and negative sequence coupled output impedance matrix in S3 specifically includes:

[0029] Based on positive-sequence voltage, negative-sequence voltage, positive-sequence current, and negative-sequence current, a coupled virtual impedance matrix is ​​established, which includes positive-sequence virtual impedance, negative-sequence virtual impedance, and positive-negative-sequence coupling compensation impedance. The positive-negative-sequence virtual impedance voltage drop is then calculated based on this matrix.

[0030] Based on the positive and negative sequence virtual impedance voltage drop, the positive and negative sequence base voltage commands output by the VSG controller are corrected to obtain the positive and negative sequence voltage reference values ​​after virtual impedance correction.

[0031] A total output impedance matrix, taking into account the VSG's inherent output impedance matrix, the coupled virtual impedance matrix, and the digital control delay, is established as the positive and negative sequence coupled output impedance matrix.

[0032] In one alternative implementation, the positive and negative sequence coupled output impedance matrix is ​​represented as:

[0033] ;

[0034] in, This is the positive and negative sequence coupling output impedance matrix. The inherent output impedance matrix is ​​determined by the VSG power loop, voltage loop, current loop, and filter when no virtual impedance is applied. This is the positive and negative order coupling virtual impedance matrix; For digital control delay stage; This is the equivalent impedance correction term caused by the time delay.

[0035] In one alternative implementation, the fault-based network stability security domain constructed in S4 is defined as the set of optimization parameters that satisfy the following inequality constraints:

[0036] ;

[0037] in, x represents the stable and secure domain of the fault network; x represents the set of fault network parameters to be optimized. The stable distance of the characteristic root of the generalized impedance. This is the minimum allowable distance for the stability of the characteristic root of the generalized impedance; This represents the peak value of the impedance closed-loop sensitivity. This represents the maximum allowable value for the peak value of the impedance closed-loop sensitivity. The maximum coupling strength is determined by the positive and negative order. This represents the maximum allowable value for positive and negative order coupling strength; The characteristic frequency shift, This represents the maximum allowable value for the characteristic frequency drift. For VSG power angle synchronization stiffness, This is the minimum allowable value for the synchronous stiffness of the VSG power angle; This represents the peak value of the fault output current. The permissible fixed current safety limit for grid-type energy storage converters is determined by the power device withstand capability, filter capacity, current protection setting value, or engineering specifications. For reactive power support error, Maximum allowable error for reactive power support.

[0038] In one optional implementation, in S5, the optimization variables of the multi-objective optimization function are:

[0039] ;

[0040] Where x is the optimization variable of the multi-objective optimization function. For positive sequence virtual resistance, For positive sequence virtual inductance, For negative sequence virtual resistance, For negative sequence virtual inductance, This is a positive and negative sequence coupling compensation resistor. For positive and negative sequence coupling compensation inductors, This refers to the virtual inertia during the fault period. The damping coefficient during the fault period. This is the fault reactive voltage support factor. The active power derating factor is the fault-related factor.

[0041] In an optional implementation, in step S6, the stable safety domain-guided multi-strategy improved snake swarm optimization algorithm includes the following steps:

[0042] An initial population is generated using low-difference sequences, and individuals in the population that do not meet the stability safety domain constraint of fault-based network construction are repaired by safety domain projection.

[0043] Define a search temperature factor related to fault severity and impedance stability margin to dynamically adjust the algorithm's global exploration and local development behavior;

[0044] Define a stable food quality function that integrates impedance stability index and objective function value to evaluate and guide the evolutionary direction of the population;

[0045] The population is divided into male and female subgroups for collaborative search, and a stable eigenvalue traction mechanism based on the stable distance gradient of the generalized impedance eigenvalue is introduced.

[0046] After individual updates, the impedance-current safety repair operator is used to constrain and repair the control parameters to ensure that the individual meets or approaches the fault network stability safety domain.

[0047] In one optional implementation, S7 specifically includes:

[0048] The virtual inertia during the fault period, the fault damping coefficient, and the active power derating coefficient in the optimal fault network control parameter combination are embedded into the corresponding control equations of the active power-frequency control loop.

[0049] The reactive power support coefficient in the optimal fault network control parameter combination is embedded into the corresponding control equation of the reactive power-voltage control loop.

[0050] The positive-sequence virtual impedance, negative-sequence virtual impedance, and positive-negative-sequence coupling compensation impedance parameters from the optimal fault network control parameter combination are input into the positive-negative-sequence virtual impedance control loop to correct the voltage command.

[0051] Preset a fixed current safety limit Set as the limiting boundary of the current vector limiting loop to constrain the magnitude of the positive and negative sequence current vectors during a fault.

[0052] In one alternative implementation, positive and negative sequence current vector circular limiting constraints are established in the current vector limiting loop, and when the current capacity is limited, current is allocated according to the allocation priority of positive sequence reactive current over positive sequence active current and positive sequence active current over negative sequence current.

[0053] Secondly, embodiments of this application provide an electronic device, including a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the method provided in embodiments of this application.

[0054] Thirdly, embodiments of this application provide a computer-readable storage medium having a computer program stored thereon, which, when executed in a computer, causes the computer to perform the method provided in embodiments of this application.

[0055] The technical solution provided in this application identifies fault types by collecting and decomposing positive and negative sequence electrical quantities, and establishes an accurate positive and negative sequence coupled output impedance model accordingly. Based on the principles of asymmetric circuits and impedance analysis, it overcomes the limitations of traditional single-sequence impedance modeling, achieving accurate characterization of system impedance interaction characteristics during faults, especially under asymmetric operating conditions, thus laying an accurate analytical foundation for subsequent stability assurance. By constructing a fault network stability and safety domain that integrates multi-dimensional constraints such as impedance stability, power angle synchronization, reactive power support, and current boundaries, and using this domain as a hard constraint, and guided by a comprehensive multi-objective function, an improved optimization algorithm is employed for joint parameter optimization. Based on the principles of multi-constraint optimization and intelligent optimization, the traditional experience-based parameter tuning is transformed into systematic optimization within a clearly defined stability boundary, ensuring the final stability. The resulting combination of control parameters can simultaneously ensure the impedance stability, power angle stability, and reactive power support capability of the system during faults in principle, while actively suppressing the fault current within the safety boundary, achieving a fundamental improvement from single current limiting to a combination of stability and current limiting. Furthermore, by embedding the optimized parameters into the VSG active-frequency loop, reactive-voltage loop, positive and negative sequence virtual impedance loop, and current vector limiting loop, the invention achieves coordinated control of overcurrent suppression and grid support capability maintenance during faults. Based on this, the invention can effectively reduce the peak fault current of grid-connected energy storage converters under typical voltage drop conditions, improve the negative sequence current suppression effect, and enhance impedance stability and operational support capability during faults. This provides a reliable control strategy and technical support for grid-connected energy storage to participate in fault ride-through and grid stability support under conditions of high proportion of new energy grid connection. Attached Figure Description

[0056] Figure 1 This is a logical schematic diagram of the fault grid control method for grid-type energy storage based on coupling impedance security domain optimization provided in the embodiments of this application;

[0057] Figure 2 A system topology diagram of a fault grid-based energy storage network control method based on positive and negative sequence coupling impedance stability security domain optimization provided in this application embodiment;

[0058] Figure 3 A flowchart illustrating the overall process of the fault grid control method for grid-type energy storage based on coupling impedance security domain optimization provided in this application embodiment;

[0059] Figure 4 This is a block diagram of voltage drop fault identification and positive / negative sequence decomposition provided in an embodiment of this application;

[0060] Figure 5 This is a block diagram of output impedance modeling considering positive and negative sequence virtual impedance and digital delay provided in an embodiment of this application;

[0061] Figure 6A schematic diagram of the stable safety domain of the fault network based on positive and negative sequence coupling impedance provided in the embodiments of this application;

[0062] Figure 7 A flowchart of the stable security domain-guided multi-strategy improved snake swarm optimization algorithm provided in the embodiments of this application;

[0063] Figure 8 VSG fault network control block diagram with optimized parameters provided in the embodiments of this application;

[0064] Figure 9 A comparison diagram of the fault current of traditional VSG control and the method of the present invention when the grid connection point voltage symmetrically drops to 0.8 pu, as provided in the embodiments of this application; Figure 9 (a) is a waveform diagram when the grid connection point voltage drops symmetrically to 0.8 pu; Figure 9 (b) The current waveform at the grid connection point when using the traditional VSG control method; Figure 9 (c) A waveform diagram of the grid connection point using the method proposed in this invention;

[0065] Figure 10 A comparison diagram of the fault current of traditional VSG control and the method of the present invention when the grid connection point voltage symmetrically drops to 0.2 pu, as provided in the embodiments of this application; Figure 10 (a) is a waveform diagram when the grid connection point voltage drops symmetrically to 0.2 pu; Figure 10 (b) The current waveform at the grid connection point when using the traditional VSG control method; Figure 10 (c) A waveform diagram of the grid connection point using the method proposed in this invention;

[0066] Figure 11 A comparison diagram of the fault current of traditional VSG control and the method of the present invention when the grid connection point asymmetric voltage drops to 0.5 pu, as provided in the embodiments of this application; Figure 11 (a) is a waveform diagram when the unbalanced voltage at the grid connection point drops to 0.5 pu; Figure 11 (b) The current waveform at the grid connection point when using the traditional VSG control method; Figure 11 (c) is a waveform diagram of the grid connection point using the method proposed in this invention. Detailed Implementation

[0067] The present application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0068] Figure 1 This is a logic diagram of the fault grid control method for grid-type energy storage based on coupling impedance security domain optimization provided in the embodiments of this application. Figure 3This is a general flowchart of the grid-type energy storage fault grid control method based on coupling impedance security domain optimization provided in the embodiments of this application. The method can be executed by the grid-type energy storage fault grid control system based on coupling impedance security domain optimization. The system can be implemented by software and / or hardware and can be configured in electronic devices such as computers.

[0069] like Figure 1 and Figure 3 As shown, the technical solution provided in this application includes the following steps:

[0070] S1. Establish a grid-connected energy storage system and voltage sag fault model, and determine the mathematical relationships between the energy storage converter output voltage, filter inductor current, point of common coupling voltage, grid current, grid equivalent impedance, voltage sag depth, positive and negative sequence disturbance components and the virtual synchronous generator (VSG) control parameters.

[0071] In some embodiments, establishing a grid-connected energy storage system and a voltage dip fault model specifically includes:

[0072] S11. Establish the system topology and basic electrical model.

[0073] Specifically, Figure 2 This application provides a system topology diagram for a grid-based energy storage fault grid control method based on positive and negative sequence coupling impedance stability security domain optimization. The invention is applied to, for example... Figure 2 The diagram shows a grid-connected energy storage system consisting of energy storage units, a bidirectional DC-DC converter, a DC bus capacitor, a voltage source energy storage converter, an AC filter, a point of common coupling (PCC), grid connection lines, and the grid's equivalent impedance. The energy storage units are connected to the DC bus via the DC-DC converter, providing DC energy support for the subsequent grid-connected converter. The DC-AC converter uses VSG control and connects to the grid via a filter resistor. Filter inductor Filter capacitor AC filters and equivalent resistance of grid-connected lines Equivalent inductance of grid-connected lines The lines are connected to the PCC and the external power grid.

[0074] In some embodiments, a DC-side model for energy storage is established. The energy storage device is connected to the DC bus via a bidirectional DC-DC converter. The DC bus capacitor is used to buffer the instantaneous difference between the energy storage-side power and the AC-side output power. Its dynamic equation is:

[0075] ;

[0076] In the formula, This refers to the DC bus capacitance, measured in F. This is the DC bus voltage, in V. The input DC current to the energy storage side is expressed in amperes (A). This is the equivalent DC current consumed on the AC side of the energy storage converter, expressed in amperes (A).

[0077] In some embodiments, a grid-type energy storage AC-side filter model is established, wherein the energy storage converter is connected to the PCC via a filter inductor and a filter capacitor. The AC-side model is as follows:

[0078] ;

[0079] In the formula, This is a filter inductance, measured in watts (H). This is the filter resistor, in Ω. This is a filter capacitor, measured in F (amperes). The output voltage vector of the energy storage converter bridge arm; The voltage vector at the point of common coupling; The current vector of the filter inductor; This is the grid-connected current vector.

[0080] In some embodiments, an equivalent impedance model of the power grid is established, and the PCC is connected to the external power grid via the line impedance, satisfying:

[0081] ;

[0082] In the formula, This is the equivalent inductance of the power grid, measured in ohms (H). This is the equivalent resistance of the power grid, measured in Ω. This represents the grid voltage vector.

[0083] S12. Establish a voltage drop fault model.

[0084] Specifically, in order to analyze the dynamics during a fault, it is necessary to model the state of the grid voltage under the fault.

[0085] In some embodiments, a symmetrical voltage dip fault model is established. When a three-phase symmetrical voltage dip occurs, the three-phase voltage of the power grid is expressed as:

[0086] ;

[0087] In the formula, This represents the time-domain vector of the three-phase voltage of the power grid under fault conditions. This refers to the three-phase symmetrical voltage drop depth; This is the rated phase voltage amplitude; This is the rated angular frequency.

[0088] In some embodiments, an asymmetrical voltage dip fault model is established. When an asymmetrical voltage dip such as a single-phase ground fault, a two-phase short circuit, or a two-phase ground fault occurs, the grid voltage is expressed as:

[0089] ;

[0090] In the formula, This represents the positive sequence voltage component of the power grid. This represents the negative sequence voltage component of the power grid. This is the inverse transformation matrix from positive and negative sequence components to the three-phase coordinates. This model shows that a fault introduces a negative sequence disturbance component.

[0091] S13. Establish a virtual synchronous generator (VSG) control model.

[0092] In some embodiments, the grid-type energy storage converter uses VSG control, and its control model includes the following steps:

[0093] The active-frequency control equations of the VSG are established. The VSG regulates the frequency and active power by simulating the rotor motion equations of a synchronous generator.

[0094] ;

[0095] In the formula, VSG virtual inertia, measured in seconds (s). The output angular frequency of the VSG is expressed in rad / s. The rated angular frequency is expressed in rad / s. This is a reference value for active power, in watts (W). The active power output of the energy storage converter is expressed in watts (W). The damping coefficient; This is the reference value for the phase angle of the VSG output voltage, in rad.

[0096] In some embodiments, active power derating control equations are established during fault periods, and adjustable parameters are introduced into the basic model to adapt to fault conditions.

[0097] ;

[0098] In the formula, This is the corrected active power reference value during the fault period; The active power derating factor for faults has the following range: ;

[0099] The active power-frequency control equation during the fault period is:

[0100] ;

[0101] In the formula, This refers to the virtual inertia during the fault period. This represents the damping coefficient during the fault.

[0102] In some embodiments, a VSG reactive power-voltage control equation is established, in which the VSG supports voltage and reactive power by adjusting the amplitude of its internal potential:

[0103] During normal operation:

[0104] ;

[0105] In the formula, This is a reference value for the amplitude of the VSG internal potential. The rated voltage amplitude is the reference value. This is the reactive power regulation coefficient; This is a reference value for reactive power. To output reactive power to the energy storage converter; This is the voltage regulation coefficient; This represents the PCC voltage amplitude.

[0106] During the outage:

[0107] ;

[0108] In the formula, Command to set the VSG internal potential amplitude during the fault period; This is the fault reactive voltage support coefficient; This represents the positive sequence voltage amplitude of the PCC.

[0109] S2. Collect the three-phase voltage and three-phase current at the point of common coupling, and decompose the three-phase voltage and three-phase current into positive and negative sequence to obtain positive sequence voltage, negative sequence voltage, positive sequence current and negative sequence current; based on the positive sequence voltage drop depth and negative sequence imbalance coefficient, identify whether the current fault is a symmetrical voltage drop condition or an asymmetrical voltage drop condition.

[0110] Specifically, such as Figure 4 As shown, the specific execution process of S2 is as follows:

[0111] S21. The system samples the three-phase instantaneous voltage and three-phase instantaneous current at the PCC in real time.

[0112] S22. Perform positive and negative sequence decomposition on the three-phase voltage and three-phase current.

[0113] In some embodiments, in order to separately process the positive-sequence component (representing fundamental energy transmission) and the negative-sequence component (caused by asymmetrical faults) under fault conditions, it is necessary to rapidly decompose the acquired three-phase signal. This embodiment adopts the instantaneous symmetrical component method, which uses a fixed transformation matrix to transform the signal in the three-phase stationary coordinate system to a coordinate system containing zero-sequence, positive-sequence, and negative-sequence components.

[0114] For the decomposition of voltage signals:

[0115] ;

[0116] in, For rotation operators, It is the zero-sequence voltage component; It is the positive sequence voltage component; It is a negative sequence voltage component; , , These are the three-phase PCC voltages, A, B, and C, respectively.

[0117] The same transformation matrix is ​​used for the decomposition of current signals:

[0118] ;

[0119] in, This is the zero-sequence current component; It is the positive sequence current component; It is the negative sequence current component; , , These are the three-phase grid-connected currents of phases A, B, and C, respectively.

[0120] S23. Calculate the fault characteristic quantities. Based on the positive and negative order components obtained from the decomposition, calculate two core quantitative indicators for fault judgment.

[0121] Calculate the positive sequence voltage drop depth:

[0122] ;

[0123] In the formula, This represents the positive-sequence voltage sag depth, used to measure the degree of decline in the grid's voltage support capacity. A larger value indicates a more severe voltage sag. This represents the positive sequence voltage amplitude of the PCC. This is the rated voltage amplitude.

[0124] Calculate the negative order imbalance coefficient:

[0125] ;

[0126] In the formula, This is a negative-sequence imbalance coefficient used to measure the degree of voltage asymmetry in the power grid; the larger the value, the stronger the voltage asymmetry. This represents the negative sequence voltage amplitude of the PCC. To prevent small positive numbers with a denominator of zero.

[0127] S24. Establish fault type criteria, compare the calculated feature quantities with preset thresholds, and thus automatically determine the fault type.

[0128] When satisfied When the voltage drops, it is determined to be a symmetrical voltage drop condition; this indicates that the overall voltage amplitude decreases, but the three phases remain basically balanced and the negative sequence component is very small.

[0129] When satisfied When the voltage drops, it is determined to be an asymmetrical voltage drop condition; this indicates that not only does the voltage amplitude decrease, but there is also a significant voltage imbalance (negative sequence component).

[0130] In the formula, A threshold for identifying voltage dips; Threshold for identifying negative order imbalance.

[0131] S3. Based on positive-sequence voltage, negative-sequence voltage, positive-sequence current and negative-sequence current, establish a positive and negative sequence coupled output impedance model for grid-type energy storage during voltage drop faults. Incorporate the inherent output impedance of VSG, positive-sequence virtual impedance, negative-sequence virtual impedance, positive and negative sequence coupled compensation impedance, filter impedance, voltage and current inner loop and digital control delay into a unified impedance modeling framework to form a positive and negative sequence coupled output impedance matrix.

[0132] Specifically, such as Figure 5 As shown, the specific execution process of S3 is as follows:

[0133] S31. Define virtual impedance and coupling compensation impedance.

[0134] In some embodiments, a positive-sequence virtual impedance is established, which is the equivalent impedance introduced in the positive-sequence control channel to adjust the positive-sequence characteristics:

[0135] ;

[0136] In the formula, This is the positive-sequence virtual impedance; This is a positive-sequence virtual resistor; is the positive-sequence virtual inductance; s is the Laplace operator.

[0137] In some embodiments, a negative-sequence virtual impedance is established, which is the equivalent impedance introduced in the negative-sequence control channel to suppress negative-sequence current.

[0138] ;

[0139] In the formula, It is a negative-sequence virtual impedance; This is a negative-sequence virtual resistor; It is a negative-sequence virtual inductor.

[0140] In some embodiments, to compensate for the unavoidable mutual influence (coupling effect) between positive and negative sequence channels, a positive and negative sequence coupling compensation impedance is established:

[0141] ;

[0142] In the formula, For positive and negative sequence coupling compensation impedance; For coupling compensation resistors; This is a coupling compensation inductor.

[0143] S32. Establish a positive and negative sequence coupling virtual impedance matrix.

[0144] The three impedances mentioned above are combined into a 2x2 matrix to uniformly describe their combined effect on positive and negative sequence voltages:

[0145] ;

[0146] This matrix defines the voltage drop equation generated by the virtual impedance element:

[0147] ;

[0148] Right now:

[0149] '

[0150] In the formula, This is the positive and negative order coupling virtual impedance matrix. This represents the positive-sequence virtual impedance voltage drop. This represents the negative-sequence virtual impedance voltage drop. It is a positive sequence current; This is the negative sequence current. The equation clearly reveals the negative sequence current. Impedance will be compensated through positive and negative sequence coupling. Affecting the positive sequence virtual impedance voltage drop ,vice versa.

[0151] In some embodiments, a correction equation for the positive and negative sequence reference voltage of the VSG is established:

[0152] ;

[0153] In the formula, , These are the positive and negative sequence base voltage commands output by the VSG controller, respectively. , These are the positive and negative sequence voltage reference values ​​after virtual impedance correction, respectively.

[0154] S33, Form a positive and negative sequence coupled output impedance matrix.

[0155] In some embodiments, the complete port impedance characteristics of the grid-type energy storage converter are jointly determined by the VSG control algorithm itself, the introduced virtual impedance, the physical filter, and the delay of the digital control system, establishing the positive and negative sequence output impedance matrix of the grid-type energy storage:

[0156] ;

[0157] in:

[0158] ;

[0159] In the formula, The output impedance is the result of the positive-sequence current disturbance being converted to the positive-sequence voltage disturbance. The output impedance is the result of a negative sequence current disturbance being converted to a negative sequence voltage disturbance. The cross-coupling impedance from negative-sequence current disturbance to positive-sequence voltage disturbance; It is the cross-coupling impedance from positive-sequence current disturbance to negative-sequence voltage disturbance.

[0160] Establish the total output impedance matrix that takes into account virtual impedance and digital delay:

[0161] ;

[0162] In the formula, The inherent output impedance matrix is ​​determined by the VSG power loop, voltage loop, current loop, and filter when no virtual impedance is applied. This is the positive and negative order coupling virtual impedance matrix; This includes the sampling, calculation, and PWM modulation delay stages; This is the equivalent impedance correction term caused by the time delay.

[0163] In some embodiments, a digital control delay model is established:

[0164] ;

[0165] In the formula, This refers to the digital control delay time, which is the equivalent delay time for sampling, calculation, and PWM execution.

[0166] S4. Based on the positive and negative sequence coupled output impedance model and the identified operating conditions, a fault network stability and safety domain is constructed. The fault network stability and safety domain includes the stability distance of the generalized impedance characteristic root, the peak value of the impedance closed-loop sensitivity, the positive and negative sequence coupling strength, the characteristic frequency drift, the VSG power angle synchronous stability constraint, the reactive power support constraint, and the fault current boundary constraint.

[0167] Specifically, such as Figure 6 As shown, the specific execution process of S4 is as follows:

[0168] S41. Define core stability and performance indicators, including the stability distance of the generalized impedance characteristic root, the peak value of the impedance closed-loop sensitivity, the positive and negative sequence coupling strength, the characteristic frequency drift, the VSG power angle synchronous stability constraint, the reactive power support constraint, and the fault current boundary constraint.

[0169] In some embodiments, the positive and negative sequence equivalent admittance matrix of the external power grid is established. :

[0170] ;

[0171] in:

[0172] ;

[0173] In the formula, The positive-sequence equivalent impedance of the external power grid; The negative sequence equivalent impedance of the external power grid; The positive-sequence equivalent admittance of the external power grid; It is the negative sequence equivalent admittance of the external power grid.

[0174] In some embodiments, a positive-to-negative order coupling impedance ratio matrix is ​​established:

[0175] ;

[0176] In the formula, This is the positive and negative sequence coupling impedance ratio matrix between grid-type energy storage and the external power grid.

[0177] In some embodiments, the generalized impedance eigenvalue stabilization distance Calculate according to the following steps:

[0178] Calculate the eigenvalues ​​of the impedance ratio matrix:

[0179] ;

[0180] In the formula, The positive and negative order coupling impedance ratio matrix The k-th eigenvalue, where j is the imaginary unit.

[0181] Calculate the stable distance of the generalized eigenvalues:

[0182] ;

[0183] In the formula, The frequency band for fault network stability analysis.

[0184] when A larger value indicates that the trajectory of the impedance ratio matrix eigenvalues ​​is closer to the critical point of the complex plane. The lower the risk of system impedance coupling instability; when If this occurs, it is determined that the current control parameters do not meet the requirements of the fault network stability and security domain.

[0185] In some embodiments, the peak impedance closed-loop sensitivity Calculate according to the following steps:

[0186] Establish the positive and negative sequence impedance closed-loop sensitivity matrix:

[0187] ;

[0188] In the formula, is the positive and negative sequence impedance closed-loop sensitivity matrix; I is the identity matrix.

[0189] Calculate the peak value of the maximum singular value :

[0190] ;

[0191] In the formula, This represents the largest singular value of the matrix.

[0192] when When the value increases, it indicates that voltage dip disturbances, fault recovery disturbances, or negative sequence disturbances are amplified more significantly through the impedance coupling path; when When the current control parameters meet the requirements for disturbance amplification and suppression during the fault period, it is determined that the current control parameters meet the requirements for disturbance amplification and suppression during the fault period.

[0193] In some embodiments, the positive and negative order coupling strengths are calculated according to the following steps:

[0194] Establish frequency domain positive and negative order coupling strength indices:

[0195] ;

[0196] In the formula, Angular frequency The positive and negative order coupling strength at the location; To prevent small positive numbers with a denominator of zero.

[0197] Establish a maximum positive and negative order coupling strength index:

[0198] ;

[0199] In the formula, The key frequency bands to focus on for fault network construction include the low-frequency power angle dynamic band, the mid-frequency impedance interaction band, and the filter resonant band.

[0200] when The larger the value, the more easily negative-order disturbances affect the positive-order network support channels; when... The current control parameters are determined to meet the positive and negative sequence decoupling stability requirements under asymmetrical voltage drop.

[0201] In some embodiments, the characteristic frequency drift amount Calculate according to the following steps:

[0202] Calculate the impedance characteristic frequency under normal operating conditions:

[0203] ;

[0204] Calculate the characteristic frequency of the impedance under fault operating conditions:

[0205] ;

[0206] Calculate the characteristic frequency shift:

[0207] ;

[0208] In the formula, This represents the characteristic frequency corresponding to the k-th impedance characteristic value trajectory under normal conditions. The characteristic frequency is the frequency corresponding to the k-th impedance characteristic value trajectory under fault conditions.

[0209] When the characteristic frequency approaches the filter resonant frequency band, the control bandwidth crossover frequency band, or the PWM delay sensitive frequency band, the optimization penalty term is increased to avoid mid-to-high frequency oscillations caused by improper selection of virtual impedance parameters.

[0210] In some embodiments, the VSG power angle synchronization stability constraint includes the following steps:

[0211] Establish the dynamic equation of VSG power angle:

[0212] ;

[0213] In the formula, The power angle is the distance between the internal potential of the VSG and the positive sequence voltage of the PCC. The rate of change of the work angle; It represents the positive sequence voltage angular frequency of the power grid.

[0214] Establish VSG synchronization stiffness index:

[0215] ;

[0216] In the formula, Synchronous stiffness during the fault; Output active power to VSG.

[0217] Establish synchronous stability constraints for the work angle:

[0218]

[0219] In the formula, This is the lower limit of synchronous stiffness; This is the upper limit of the power angle offset; This is the upper limit of the rate of change of the work angle.

[0220] When the above constraints are not met, adjustments can be made. This prevents the VSG from continuously accelerating during voltage drops and from losing synchronism after fault clearing.

[0221] In some embodiments, reactive power support constraints are used to ensure that, during a fault, grid-connected energy storage can provide reactive current on demand to support grid voltage recovery. The reactive power support error is required. For reactive power support error, ,in, This represents the maximum allowable error for reactive power support. This is the positive-sequence reactive current command during the fault period. This represents the actual positive-sequence reactive current.

[0222] In some embodiments, fault current boundary constraints are used to ensure that the current flowing through the power devices and AC filters of the grid-connected energy storage converter does not exceed a fixed safety boundary, requiring the peak three-phase output current to meet the following during a voltage dip fault: ≤ ,in, This represents the peak value of the three-phase output current during the fault period. The upper limit of the fixed current allowed for grid-type energy storage converters is determined by the withstand capability of power devices, filter capacity, current protection settings, or engineering specifications.

[0223] S42. Integrate all the above constraints to establish a fault-tolerant network stability security domain:

[0224] ;

[0225] In the formula, x represents the stable and secure domain of the fault network; x represents the set of fault network parameters to be optimized. The stable distance of the characteristic root of the generalized impedance. This is the minimum allowable distance for the stability of the characteristic root of the generalized impedance; This represents the peak value of the impedance closed-loop sensitivity. This represents the maximum allowable value for the peak value of the impedance closed-loop sensitivity. The maximum coupling strength is determined by the positive and negative order. This represents the maximum allowable value for positive and negative order coupling strength. The characteristic frequency shift, This represents the maximum allowable value for the characteristic frequency drift. For VSG power angle synchronization stiffness, This is the minimum allowable value for the synchronous stiffness of the VSG power angle; This represents the peak value of the fault output current. The upper limit of the fixed current allowed for grid-type energy storage converters; For reactive power support error, Maximum allowable error for reactive power support.

[0226] S5. Establish a multi-objective optimization function with the goals of fault overcurrent suppression, impedance stability improvement, positive and negative sequence coupling reduction, reactive power support enhancement, power angle offset suppression, and parameter coordination improvement.

[0227] Specifically, establishing a multi-objective optimization function includes the following steps:

[0228] S51. Define the optimization variable vector:

[0229] ;

[0230] In the formula, This is a positive-sequence virtual resistor; This is a positive-sequence virtual inductance; This is a negative-sequence virtual resistor; It is a negative-sequence virtual inductor; For positive and negative sequence coupling compensation resistors; This is a positive and negative sequence coupling compensation inductor; This refers to the virtual inertia during the fault period; The damping coefficient during the fault; This is the fault reactive voltage support coefficient; The active power derating factor is the factor for faults.

[0231] S52. Establish the comprehensive objective function:

[0232] ;

[0233] In the formula, J(x) is the comprehensive objective function; This is an indicator of fault overcurrent. The characteristic root stability index of impedance; This refers to the peak sensitivity index; It is an index of positive and negative order coupling strength; This is a characteristic frequency drift index; For the synchronous stability index of the power angle; For reactive power support error indicators; For parameter smoothness index; To constrain penalty items; The weight coefficients are defined as follows: all eight weight coefficients are greater than or equal to 0 and the sum of the eight weight coefficients is 1.

[0234] S53. Define the sub-functions of the comprehensive objective function.

[0235] In some embodiments, a fault overcurrent index is established:

[0236] ;

[0237] In the formula, This is an indicator of fault overcurrent. This represents the peak value of the three-phase output current during the fault period. The upper limit of the fixed current allowed for grid-type energy storage converters; Rated current; , These are the weighting coefficients.

[0238] in:

[0239] .

[0240] In some embodiments, an impedance eigenvalue stability index is established:

[0241] ;

[0242] In the formula, The stable distance of the characteristic root of the generalized impedance.

[0243] In some embodiments, a peak sensitivity metric is established:

[0244] ;

[0245] In the formula, This represents the peak value of the impedance closed-loop sensitivity.

[0246] In some embodiments, a positive-negative order coupling strength index is established:

[0247] ;

[0248] In the formula, This represents the maximum coupling strength in both positive and negative order.

[0249] In some embodiments, a characteristic frequency drift index is established:

[0250] .

[0251] In some embodiments, a power angle synchronization stability index is established:

[0252] .

[0253] In some embodiments, a reactive power support error index is established:

[0254] ;

[0255] In the formula, This is the positive-sequence reactive current command during the fault period; This represents the actual positive-sequence reactive current.

[0256] In some embodiments, a parameter smoothness index is established:

[0257] ;

[0258] In the formula, This refers to the m-th control parameter in the k-th control cycle; These are the control parameters corresponding to the previous control cycle.

[0259] S54. Establish constraint penalty terms to force the optimization results to meet the hard constraints of the stable safe region:

[0260] ;

[0261] In the formula, ~ This is the penalty coefficient.

[0262] S6. The snake swarm optimization algorithm is improved by using a stable safety domain-guided multi-strategy algorithm. The multi-objective optimization function is used as the optimization objective and the stable safety domain of fault network construction is used as the constraint condition. The positive-sequence virtual impedance, negative-sequence virtual impedance, positive and negative sequence coupling compensation impedance, virtual inertia during the fault, fault damping coefficient, reactive power support coefficient and active power derating coefficient are jointly optimized to obtain the optimal combination of fault network construction control parameters.

[0263] For details, see Figure 7 The specific implementation process of S6 is as follows:

[0264] An improved snake swarm optimization algorithm guided by a stable safety domain is constructed. This algorithm, based on the traditional snake swarm optimization algorithm, incorporates low-difference safety domain initialization, a fault severity temperature factor, an impedance-stable food quality function, male-female dual-swarm collaborative search, stable eigenvalue traction, elite covariance learning, an impedance-current safety repair operator, and a stagnation regeneration mechanism to form an improved intelligent optimization algorithm. Specifically, it includes the following steps:

[0265] S61. Define the encoding method for individual snake groups.

[0266] Each individual snake in the swarm represents a set of fault network control parameters to be optimized:

[0267] ;

[0268] In the formula, Let i be the position of the i-th individual snake in the swarm. N represents the snake population size. These are the positive-sequence virtual resistance and positive-sequence virtual inductance corresponding to the i-th individual snake group; These are the negative-order virtual resistance and negative-order virtual inductance corresponding to the i-th individual snake group; These are the positive and negative sequence coupling compensation resistors and coupling compensation inductors corresponding to the i-th individual snake group; Let be the virtual inertia during the fault period corresponding to the i-th individual snake in the swarm; The fault damping coefficient is the fault damping coefficient corresponding to the i-th individual snake in the swarm. The fault reactive voltage support coefficient is the fault reactive voltage support coefficient corresponding to the i-th individual in the snake swarm. Let be the fault-based active power reduction coefficient corresponding to the i-th individual in the snake swarm.

[0269] S62. Improvement strategies include:

[0270] In some embodiments, a low-difference stable security domain initialization mechanism is constructed:

[0271] To ensure the quality of the search starting point and to explore the space uniformly, Sobol low-difference sequences are used to generate initial candidate individuals:

[0272] ;

[0273] In the formula, Let be the initial position of the i-th individual snake in the swarm; , These are the lower and upper bounds of the optimization variables, respectively. Let be the vector of the i-th Sobol low-difference sequence.

[0274] Perform a stable security domain feasibility assessment for each initial individual:

[0275] ;

[0276] If not satisfied, perform security domain projection repair:

[0277] ;

[0278] In the formula, To construct a stable and secure domain for fault-tolerant network construction.

[0279] Constructing a Feasibility Function for a Stable and Secure Domain

[0280] ;

[0281] In the formula, Let be the feasibility flag for the stable safety domain of the i-th individual snake group. When When, it indicates that the individual satisfies the fault network stability security domain constraint; when When this occurs, it indicates that the individual does not meet constraints such as impedance stability, current limiting, power angle synchronization, or reactive power support.

[0282] In some embodiments, a fault severity temperature factor is constructed:

[0283] Define the fault severity factor:

[0284] ;

[0285] In the formula, Let be the fault severity factor corresponding to the i-th individual snake group; This represents the positive sequence voltage drop depth. It is a negative order imbalance coefficient; The peak fault current corresponding to the i-th individual snake group; Rated current; is the stable distance of the generalized impedance eigenvalue corresponding to the i-th individual snake in the swarm; The peak value of the impedance closed-loop sensitivity corresponding to the i-th individual snake in the swarm; The maximum coupling strength in positive and negative order corresponding to the i-th individual in the snake group; This is the weighting coefficient for the severity of the fault; To prevent extremely small positive numbers with a denominator of zero.

[0286] Define the snake swarm search temperature factor:

[0287] ;

[0288] In the formula, For the i-th individual in the i-th... Search temperature in the next iteration; , These are the minimum and maximum search temperatures, respectively. This represents the maximum number of iterations. The temperature decay index; This is the fault severity adjustment factor.

[0289] The more severe the fault or the lower the impedance stability margin, Increased size allows the snake swarm to maintain a strong global search capability; when the fault is minor and individuals are close to the stable safe zone... This reduces the size of the algorithm, causing it to switch to a local fine-grained search.

[0290] In some embodiments, an impedance-stable food quality function is constructed:

[0291] Combining food quality with fault network stability in the snake swarm optimization algorithm, the stable food quality of the i-th snake individual is defined as:

[0292] ;

[0293] In the formula, The stable food quality for the i-th individual snake in the group; Let i be the comprehensive objective function corresponding to the i-th individual snake in the swarm; Let be the power angle synchronization stability index corresponding to the i-th individual snake group; Let be the reactive power support error index corresponding to the i-th individual snake group; To stabilize the weighting coefficient of food quality.

[0294] when The larger the value, the better the overall performance of the individual in terms of overcurrent suppression, impedance stability, disturbance amplification suppression, positive and negative sequence coupling reduction, and power angle stability.

[0295] Determine the globally optimal food location:

[0296] The individual with the highest stable food quality in the current population that satisfies the stable safety region constraint is defined as the globally optimal food location:

[0297] ;

[0298] In the formula, This represents the globally optimal food location in the k-th iteration.

[0299] In some embodiments, establishing a male-female dual-group cooperative search mechanism includes:

[0300] The snake population was divided into male and female subgroups:

[0301] ;

[0302] In the formula, O represents the total snake population; It is a male subgroup; It is a female subgroup.

[0303] The male subgroup is used to expand the search range of impedance parameters, focusing on searching candidate regions that meet the requirements of current limiting and impedance stability distance; the female subgroup is used for local fine-tuning search, focusing on optimizing reactive power support error, positive and negative sequence coupling strength and parameter smoothness.

[0304] Establish a global exploration and update mechanism:

[0305] Perform a global exploration update when food quality is insufficient or the search temperature is high:

[0306] ;

[0307] In the formula, These are randomly selected individuals from the snake population. This represents the current individual's reverse learning position; Let Lévy be the flight disturbance vector; For the exploration coefficient; , It is a random number in the range [0,1].

[0308] The reverse learning location is:

[0309] ;

[0310] Establish a food proximity update mechanism:

[0311] When food quality is high and the search temperature decreases, individual snakes tend to move closer to the globally optimal food location:

[0312] ;

[0313] in:

[0314] ;

[0315] In the formula, For the first Food proximity coefficient for each individual; , These are the minimum and maximum convergence coefficients, respectively; The random perturbation step size; It is a random perturbation vector; This is the step size decay coefficient.

[0316] Establish a male-female collaborative mating search mechanism:

[0317] Individuals with consistently high food quality were selected from both the male and female populations. and Generate collaborative search candidate solutions:

[0318] ;

[0319] In the formula, Candidate solutions generated by male-female cooperative search; It is the p-th preferred individual in the male subgroup; It is the qth preferred individual in the female subgroup; The co-mating coefficient has a range of values ​​of [value range missing]. .

[0320] In some embodiments, a stable eigenvalue traction mechanism is established:

[0321] To keep the search direction away from the critical stability boundary of impedance coupling, a generalized impedance eigenvalue stability distance traction term is introduced during the individual update process:

[0322] ;

[0323] In the formula, The traction coefficient is a stable characteristic root. This is the gradient of the stable distance of the generalized impedance eigenvalue with respect to the optimization variables.

[0324] The gradient is calculated using a finite difference approximation:

[0325] ;

[0326] In the formula, Let n be the optimization variable for the i-th individual; Let n be the small perturbation of the nth variable.

[0327] The stable and safe domain-guided multi-strategy improved snake swarm optimization algorithm also includes elite covariance learning, impedance-current safety repair operator, and stagnation regeneration mechanism, specifically including the following steps:

[0328] In some embodiments, an elite covariance learning matrix is ​​constructed:

[0329] Select the top stable food quality rankings in the current iteration Individuals considered elite individuals have the following mean:

[0330] ;

[0331] The elite covariance matrix is:

[0332] ;

[0333] In the formula, The number of elite individuals; For the qth elite individual; is the regularization coefficient; I is the identity matrix.

[0334] Local learning perturbations are generated based on the elite covariance matrix:

[0335] ;

[0336] In the formula, It is a standard normal random vector.

[0337] In some embodiments, constructing the impedance-current safety repair factor includes:

[0338] To prevent updated individuals from violating impedance stability, current limiting, and positive / negative sequence coupling constraints, a safety repair factor is defined:

[0339] ;

[0340] In the formula, Let be the safety repair coefficient for the i-th individual.

[0341] The virtual impedance parameter is corrected based on the safety repair factor:

[0342] ;

[0343] In the formula, , , , These are the safety repair steps for positive-sequence virtual resistance, positive-sequence virtual inductance, negative-sequence virtual resistance, and negative-sequence virtual inductance, respectively.

[0344] Adjust VSG fault control parameters based on safety repair factor:

[0345] ;

[0346] In the formula, This is the fault damping repair step size;

[0347] Construct a parameter boundary repair mechanism

[0348] Boundary fixes are performed on the updated individuals:

[0349] ;

[0350] In the formula, This represents the updated value of the m-th variable for the i-th individual; , These are the lower and upper limits of the m-th variable, respectively.

[0351] In some embodiments, constructing a stalled regeneration mechanism includes:

[0352] When continuous If the global optimal objective function fails to improve in the next iteration, the population search is deemed to have stagnated, and stagnant regeneration guided by the stability safety region is executed.

[0353] ;

[0354] In the formula, These are new candidate solutions generated after stagnation and regeneration. The current globally optimal food position; Individuals selected randomly; The stagnation regeneration coefficient; The fault severity factor corresponding to the globally optimal food location; T is the stabilizing traction coefficient for regeneration. s The number of iterations is used to determine when the system stalls.

[0355] Establish convergence criteria for the algorithm:

[0356] Iteration stops when any of the following conditions are met:

[0357]

[0358] or

[0359] And continue M c The next iteration;

[0360] In the formula, This represents the maximum number of iterations. This represents the globally optimal objective function value at the k-th iteration. M is the convergence threshold; c The number of times to maintain convergence.

[0361] S63, Output the optimal combination of control parameters.

[0362] When the maximum number of iterations is reached, or M consecutive iterations are reached... c The algorithm terminates when the change in the global optimal objective function is less than the convergence threshold, and finally outputs the optimal combination of fault network control parameters that satisfies the stability and safety domain constraints of the fault network:

[0363] ;

[0364] In the formula, To improve the optimal combination of fault network control parameters output by the stable and safe domain-guided multi-strategy snake swarm optimization algorithm.

[0365] S7. The optimal fault grid-connection control parameter combination is embedded into the active-frequency control loop, reactive-voltage control loop, positive and negative sequence virtual impedance control loop and current vector limiting loop of the virtual synchronous generator respectively to generate the three-phase voltage reference value of the grid-connected energy storage converter during voltage drop faults, so as to perform fault grid-connection control.

[0366] For details, see Figure 8 The specific implementation process of S7 is as follows:

[0367] S71. The optimal fault network control parameter combination is embedded into each control loop of the virtual synchronous generator.

[0368] In some embodiments, embedding the optimization parameters into the VSG control loop includes the following steps:

[0369] Embedded optimized VSG fault control parameters:

[0370] ;

[0371] Establish the fault active power-frequency control equation after embedding optimized parameters:

[0372] ;

[0373] In the formula, This refers to the optimized virtual inertia during the fault period. The optimized fault damping coefficient; This is the optimized active power derating factor.

[0374] Establish the fault reactive power-voltage control equation with embedded optimized parameters:

[0375] ;

[0376] In the formula, This is the optimized fault reactive voltage support factor.

[0377] Embedded optimized positive and negative sequence virtual impedance parameters:

[0378] ;

[0379] In the formula, This is the optimized positive and negative sequence coupling virtual impedance matrix.

[0380] Generate positive and negative sequence voltage reference values:

[0381] ;

[0382] After completing parameter embedding and command correction, the corrected positive and negative sequence voltage reference values ​​are synthesized into a voltage reference signal in a three-phase stationary coordinate system through an inverse transformation matrix:

[0383] ;

[0384] In the formula, The three-phase voltage reference value is used as the final input to the PWM modulation stage. The generated three-phase voltage reference value is sent to the voltage and current dual closed-loop controller, where it is compared with the actual capacitor voltage and inductor current and adjusted by PI to generate a modulation wave. The modulation wave passes through the pulse width modulation (PWM) stage to generate a series of switching transistor drive signals, which ultimately control the power devices (such as IGBTs) of the grid-type energy storage converter to operate, causing them to output the desired voltage and current waveforms.

[0385] In some embodiments, current vector limiting and reactive power support allocation include the following steps:

[0386] Establish circular limiting constraints for positive and negative sequence current vectors:

[0387] ;

[0388] In the formula, , These are the positive sequence d-axis and q-axis currents, respectively; , These are the negative sequence d-axis and q-axis currents, respectively; This refers to the maximum safe fixed current limit allowed for grid-type energy storage converters.

[0389] Establish fault positive sequence reactive current command:

[0390] ;

[0391] In the formula, This is the reactive current support factor; This represents the positive sequence voltage drop depth. This is the rated current.

[0392] Establish current allocation priority:

[0393] ;

[0394] When current capacity is limited, priority is given to ensuring positive-sequence reactive current, followed by positive-sequence active current, and finally suppressing negative-sequence current. During a fault, the strategy prioritizes ensuring sufficient positive-sequence reactive current to support grid voltage recovery (a core function of the grid); secondly, it maintains active power transmission as much as possible; and finally, it utilizes remaining current capacity to suppress negative-sequence current and improve three-phase imbalance. This strategy is implemented through a dynamic current command allocation algorithm.

[0395] Establish the formula for allocating residual current capacity:

[0396] When satisfied season ;

[0397] Residual current capacity I res for:

[0398] ;

[0399] The positive sequence active current command is:

[0400] ;

[0401] In the formula, This is a limiting function; This is a reference value for active power during the fault period; This represents the positive sequence voltage amplitude of the PCC.

[0402] Next, simulation verification will be performed. Figure 2The simulation model is established based on the topology of the grid-type energy storage fault network system shown. The simulation system mainly includes energy storage devices, a bidirectional DC-DC converter, a DC bus, a grid-type energy storage converter, an AC filter, line impedance, a PCC (Point of Common Coupling), and an equivalent grid model. The energy storage devices are connected to the DC bus via the bidirectional DC-DC converter, providing DC energy support for the grid-type energy storage converter. The grid-type energy storage converter uses VSG control and is connected to the grid side via a filter and line impedance to simulate the fault network operation process during voltage dip faults.

[0403] During the simulation, the rated DC voltage of the energy storage device was set to 600V, and the grid-connected energy storage converter maintained normal grid-connected operation before the fault. The total simulation time was set to 5.25s, and a voltage dip fault was applied at 5s to simulate the fault current response characteristics of the grid-connected energy storage converter under grid fault conditions. To verify the applicability of the method of the present invention under different fault types, three typical operating conditions were set in the simulation: symmetrical voltage dip to 0.2pu, symmetrical voltage dip to 0.8pu, and asymmetrical voltage dip to 0.5pu. The first two operating conditions were used to verify the ability of the method of the present invention to suppress the peak current at the grid connection point under symmetrical faults of different dip depths, and the third operating condition was used to verify the suppression effect of the method of the present invention on negative sequence disturbances and three-phase current imbalance under asymmetrical voltage dip conditions.

[0404] In the comparative verification, two sets of simulations were set up, one using the traditional VSG control method and the other using the control method proposed in this invention. The traditional VSG control method uses fixed VSG control parameters and a conventional current limiting strategy; the method of this invention constructs a multi-objective optimization function based on the stable safety domain of positive and negative sequence coupled impedance, and uses a stable safety domain-guided multi-strategy improved snake swarm optimization algorithm to jointly optimize the positive and negative sequence virtual impedance, coupled compensation impedance, virtual inertia during VSG faults, damping coefficient, reactive power support coefficient, and active power derating coefficient. Subsequently, the control system embeds the optimized fault-based grid control parameters into the VSG active power-frequency control loop, reactive power-voltage control loop, positive and negative sequence virtual impedance control loop, and current vector limiting loop, thereby generating three-phase voltage reference values ​​during faults and driving the grid-type energy storage converter to operate accordingly.

[0405] Figure 9 The grid connection point current is compared between the traditional VSG control method and the method proposed in this invention when the grid connection point voltage symmetrically drops to 0.8 pu. Figure 9 (a) is a waveform diagram when the grid connection point voltage drops symmetrically to 0.8 pu; Figure 9 (b) The current waveform at the grid connection point when using the traditional VSG control method; Figure 9(c) The current waveform at the grid connection point using the method proposed in this invention is shown in the figure. As can be seen from the figure, after a voltage drop, the grid connection point current exhibits a significant transient surge under traditional VSG control, with a maximum current peak of approximately 3.8 pu. However, after adopting the method proposed in this invention, the transient overcurrent at the moment of fault is significantly suppressed, and the maximum current peak is reduced to approximately 2.5 pu. This result indicates that the method of this invention can effectively reduce the fault current surge under shallow symmetrical voltage drop conditions and improve the fault current limiting capability of grid-connected energy storage converters.

[0406] Figure 10 The grid connection point current is compared between the traditional VSG control method and the method proposed in this invention when the grid connection point voltage symmetrically drops to 0.2 pu. Figure 10 (a) is a waveform diagram when the grid connection point voltage drops symmetrically to 0.2 pu; Figure 10 (b) The current waveform at the grid connection point when using the traditional VSG control method; Figure 10 (c) The current waveform at the grid connection point using the method proposed in this invention is shown in the figure. As can be seen from the figure, under deep voltage dip conditions, the traditional VSG control method generates a large transient current surge at the moment of fault occurrence, with the maximum current peak reaching 6.1 pu. However, after adopting the method proposed in this invention, the overcurrent peak is significantly suppressed, and the maximum current peak is reduced to approximately 3.1 pu. This result demonstrates that the method of this invention can effectively limit the fault current amplitude under deep symmetrical voltage dip conditions, avoiding excessive transient surge current generated by the grid-connected energy storage converter due to its low output impedance voltage source characteristics.

[0407] Figure 11 The results show a comparison of the grid connection point current when the asymmetrical voltage drop at the grid connection point is 0.5 pu, using the traditional VSG control method and the method proposed in this invention. Figure 11 (a) is a waveform diagram when the unbalanced voltage at the grid connection point drops to 0.5 pu; Figure 11 (b) The current waveform at the grid connection point when using the traditional VSG control method; Figure 11 (c) The current waveform at the grid connection point using the method proposed in this invention is shown in the figure. As can be seen from the figure, after the asymmetrical voltage drop occurs, the three-phase current under traditional VSG control exhibits significant imbalance and transient impact. The maximum peak current of the faulty phase is approximately 4.46 pu, and the maximum peak current of the non-faulty phase is approximately 3.77 pu. However, after using the method proposed in this invention, the maximum peak current of the faulty phase is reduced to approximately 2.75 pu, and the maximum peak current of the non-faulty phase is reduced to approximately 2.14 pu. This result indicates that the method of this invention can not only suppress the overcurrent of the faulty phase under asymmetrical faults, but also reduce the current impact of the non-faulty phases, thus improving the three-phase current imbalance problem during asymmetrical voltage drops.

[0408] Figures 9 to 11 The fault current suppression effect of the method proposed in this invention under different voltage dip conditions was verified. Simulation results show that, compared with the traditional VSG control method, the fault grid construction control method for grid-type energy storage based on positive and negative sequence coupling impedance stability safety domain optimization proposed in this invention can significantly reduce the peak current at the grid connection point under both symmetrical and asymmetrical voltage dip conditions, weaken the transient impact at the moment of fault occurrence, and improve the current imbalance problem under asymmetrical faults. This verifies the effectiveness and applicability of the method proposed in the fault grid construction process of grid-type energy storage.

[0409] The core objective of this invention is to improve upon a series of technical shortcomings in existing grid-connected energy storage VSG fault network control, specifically including: insufficient characterization of the positive and negative sequence coupling mechanism under voltage dip faults; reliance on empirical tuning of traditional virtual impedance parameters; difficulty in coordinating fault current limiting and impedance stability; insufficient search accuracy and global optimization capability of parameter optimization algorithms; and the failure of the optimization objective function to fully reflect multi-dimensional performance requirements such as fault current suppression, reactive power support, negative sequence suppression, and power angle stability. In existing control methods, when a symmetrical or asymmetrical voltage dip occurs in the grid, the grid-connected energy storage converter, exhibiting low output impedance voltage source characteristics, is prone to generating large current surges at the moment of fault. Simultaneously, under asymmetrical dip conditions, negative sequence voltage and current affect the positive sequence network support capability through power calculation, voltage-current inner loop, virtual impedance link, and current limiting link, leading to three-phase current imbalance, negative sequence current accumulation, and a decrease in impedance coupling stability margin. To address these problems, this invention proposes a grid-connected energy storage fault network control method based on coupling impedance safety domain optimization. This method establishes a grid-connected energy storage system and a voltage dip fault model. It constructs a model of the positive and negative sequence coupled output impedance, including the inherent output impedance of the VSG, positive and negative sequence virtual impedance, positive and negative sequence coupling compensation impedance, filter impedance, and digital control delay. This model accurately characterizes the port impedance characteristics of the grid-connected energy storage converter and the interaction between positive and negative sequence channels during a fault. Furthermore, it constructs a fault-based grid stability security domain, including the generalized impedance characteristic root stability distance, impedance closed-loop sensitivity peak, positive and negative sequence coupling strength, characteristic frequency drift, power angle synchronous stability constraint, reactive power support constraint, and fault current boundary constraint. Based on this, a multi-objective optimization function is established that considers fault overcurrent suppression, impedance stability improvement, positive and negative sequence coupling weakening, reactive power support enhancement, power angle offset suppression, and parameter coordination improvement. A stability security domain-guided multi-strategy improved snake swarm optimization algorithm is used to jointly optimize the positive and negative sequence virtual impedance, positive and negative sequence coupling compensation impedance, virtual inertia of the VSG during a fault, damping coefficient, reactive power support coefficient, and active power derating coefficient. This strategy effectively suppresses transient overcurrents caused by voltage dip faults while ensuring the voltage source support characteristics of grid-connected energy storage. It also reduces the coupling effect of negative-sequence components on the positive-sequence grid support channel under asymmetrical faults, fundamentally improving impedance stability and current safety margin during faults. Ultimately, this scheme not only significantly reduces the peak fault current at the grid connection point under symmetrical and asymmetrical voltage dip conditions, improving three-phase current imbalance and fault transient oscillations, but also balances reactive power voltage support capability with VSG power angle stability. This provides reliable technical support for grid-connected energy storage to participate in fault ride-through, grid support, and stable system operation under conditions of high-proportion renewable energy grid connection.

[0410] This application also provides an electronic device, including: at least one processor; a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to execute any one of the grid-type energy storage fault grid-building control methods based on coupling impedance security domain optimization.

[0411] This application also proposes a computer storage medium storing a computer program, which, when executed by a processor, implements any one of the following: a fault-based grid-connected energy storage grid control method based on coupling impedance security domain optimization.

[0412] Computer storage media may be simply referred to as media. Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Dual Data SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchlink DRAM (SLDRAM), Rambus Direct RAM (RDRAM), Direct Memory Bus Dynamic RAM (DRDRAM), and Memory Bus Dynamic RAM (RDRAM). The various embodiments described in this specification are presented in a progressive manner, with reference allowed to each other for similar or identical parts. Each embodiment focuses on describing the differences from other embodiments. In particular, the embodiments for apparatuses, devices, and non-volatile computer storage media are described simply because they are substantially similar to the method embodiments; relevant details can be found in the descriptions of the method embodiments.

[0413] The above embodiments are merely illustrative examples and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations. However, obvious variations or modifications derived therefrom are still within the scope of protection of this application.

Claims

1. A fault-based grid control method for grid-type energy storage based on coupling impedance security domain optimization, characterized in that, include: S1. Establish a grid-connected energy storage system and voltage sag fault model, and determine the mathematical relationship between the output voltage of the energy storage converter, the filter inductor current, the point of common coupling voltage, the grid current, the grid equivalent impedance, the voltage sag depth, the positive and negative sequence disturbance components and the control parameters of the virtual synchronous generator (VSG). S2. Collect the three-phase voltage and three-phase current of the common connection point, and decompose the three-phase voltage and three-phase current into positive and negative sequence to obtain positive sequence voltage, negative sequence voltage, positive sequence current and negative sequence current; based on the positive sequence voltage drop depth and negative sequence imbalance coefficient, identify whether the current fault is a symmetrical voltage drop condition or an asymmetrical voltage drop condition. S3. Based on the positive-sequence voltage, negative-sequence voltage, positive-sequence current and negative-sequence current, establish a positive and negative sequence coupled output impedance model for grid-type energy storage during voltage drop faults. Incorporate the inherent output impedance of VSG, positive-sequence virtual impedance, negative-sequence virtual impedance, positive and negative sequence coupled compensation impedance, filter impedance, voltage and current inner loop and digital control delay into a unified impedance modeling framework to form a positive and negative sequence coupled output impedance matrix. S4. Based on the positive and negative sequence coupled output impedance model and the identified operating conditions, a fault network stability and safety domain is constructed. The fault network stability and safety domain includes the generalized impedance characteristic root stability distance, impedance closed-loop sensitivity peak value, positive and negative sequence coupling strength, characteristic frequency drift, VSG power angle synchronous stability constraint, reactive power support constraint, and fault current boundary constraint. S5. Establish a multi-objective optimization function with the goals of fault overcurrent suppression, impedance stability improvement, positive and negative sequence coupling reduction, reactive power support enhancement, power angle offset suppression, and parameter coordination improvement. S6. Using the stable safety domain-guided multi-strategy improved snake swarm optimization algorithm, with the multi-objective optimization function as the optimization objective and the fault network stable safety domain as the constraint, the joint parameter optimization of positive sequence virtual impedance, negative sequence virtual impedance, positive and negative sequence coupling compensation impedance, virtual inertia during fault, fault damping coefficient, reactive power support coefficient and active power derating coefficient is performed to obtain the optimal combination of fault network control parameters. S7. The optimal fault grid control parameter combination is embedded into the active-frequency control loop, reactive-voltage control loop and positive-negative sequence virtual impedance control loop of the virtual synchronous generator, and combined with the preset fixed current safety upper limit applied to the current vector limiting loop to generate the three-phase voltage reference value of the grid-type energy storage converter during the voltage drop fault, so as to perform fault grid control.

2. The method according to claim 1, characterized in that, S1 specifically includes: Establish a DC-side model for energy storage, an AC-side filter model for grid-type energy storage, and a grid equivalent impedance model; Establish symmetrical voltage dip fault models and asymmetrical voltage dip fault models; A VSG control model is established, which includes active-frequency control equations, reactive-voltage control equations, active power derating control equations during faults, and reactive power voltage support equations during faults.

3. The method according to claim 1, characterized in that, In step S2, identifying the fault type based on the positive-sequence voltage drop depth and the negative-sequence imbalance coefficient includes: Calculate the positive sequence voltage drop depth: ； in, This represents the positive sequence voltage drop depth. The positive sequence voltage magnitude at the point of common coupling; This is the rated voltage amplitude; Calculate the negative order imbalance coefficient: ； in, It is a negative order imbalance coefficient. The negative sequence voltage amplitude at the point of common coupling; To prevent small positive numbers with a denominator of zero; like and If so, it is determined to be a symmetrical voltage drop condition; like and If the voltage drops, it is determined to be an unbalanced voltage drop condition; among which, A threshold for identifying voltage dips; Threshold for identifying negative order imbalance.

4. The method according to claim 1, characterized in that, The step of forming the positive and negative sequence coupled output impedance matrix in S3 specifically includes: Based on the positive-sequence voltage, negative-sequence voltage, positive-sequence current, and negative-sequence current, a coupled virtual impedance matrix is ​​established, which includes positive-sequence virtual impedance, negative-sequence virtual impedance, and positive-negative-sequence coupling compensation impedance. The positive-negative-sequence virtual impedance voltage drop is then calculated based on this matrix. Based on the positive and negative sequence virtual impedance voltage drop, the positive and negative sequence base voltage commands output by the VSG controller are corrected to obtain the positive and negative sequence voltage reference values ​​after virtual impedance correction. A total output impedance matrix, taking into account the VSG inherent output impedance matrix, the coupled virtual impedance matrix, and the digital control delay, is established as the positive and negative sequence coupled output impedance matrix.

5. The method according to claim 4, characterized in that, The positive and negative sequence coupling output impedance matrix is ​​represented as follows: ； in, This is the positive and negative sequence coupling output impedance matrix. The inherent output impedance matrix is ​​determined by the VSG power loop, voltage loop, current loop, and filter when no virtual impedance is applied. This is the positive and negative order coupling virtual impedance matrix; For digital control delay stage; This is the equivalent impedance correction term caused by the time delay.

6. The method according to claim 1, characterized in that, The fault-based network stability and security domain constructed in S4 is defined as the set of optimization parameters that satisfy the following inequality constraints: ； in, x represents the stable and secure domain of the fault network; x represents the set of fault network parameters to be optimized. The stable distance of the characteristic root of the generalized impedance. This is the minimum allowable distance for the stability of the characteristic root of the generalized impedance; This represents the peak value of the impedance closed-loop sensitivity. This represents the maximum allowable value for the peak value of the impedance closed-loop sensitivity. The maximum coupling strength is determined by the positive and negative order. This represents the maximum allowable value for positive and negative order coupling strength. The characteristic frequency shift, This represents the maximum allowable value for the characteristic frequency drift. For VSG power angle synchronization stiffness, This is the minimum allowable value for the synchronous stiffness of the VSG power angle; This represents the peak value of the fault output current. The upper limit of the fixed current allowed for grid-type energy storage converters; For reactive power support error, Maximum allowable error for reactive power support.

7. The method according to claim 1, characterized in that, In step S5, the optimization variables of the multi-objective optimization function are: ； Where x is the optimization variable of the multi-objective optimization function. For positive sequence virtual resistance, For positive sequence virtual inductance, For negative sequence virtual resistance, For negative sequence virtual inductance, This is a positive and negative sequence coupling compensation resistor. For positive and negative sequence coupling compensation inductors, This refers to the virtual inertia during the fault period. The damping coefficient during the fault period. This is the fault reactive voltage support factor. The active power derating factor is the factor for faults.

8. The method according to claim 1, characterized in that, In step S6, the stable safety domain-guided multi-strategy improved snake swarm optimization algorithm is used, which includes the following steps: An initial population is generated using low-difference sequences, and individuals in the population that do not satisfy the stability safety domain constraint of the fault network are repaired by safety domain projection. Define a search temperature factor related to fault severity and impedance stability margin to dynamically adjust the algorithm's global exploration and local development behavior; Define a stable food quality function that integrates impedance stability index and objective function value to evaluate and guide the evolutionary direction of the population; The population is divided into male and female subgroups for collaborative search, and a stable eigenvalue traction mechanism based on the stable distance gradient of the generalized impedance eigenvalue is introduced. After individual updates, the impedance-current safety repair operator is used to constrain and repair the positive and negative sequence virtual impedance, positive and negative sequence coupling compensation impedance, and VSG fault control parameters, so that the updated individual satisfies the fault network stability safety domain and the peak value of the fault output current does not exceed the fixed current safety limit allowed by the grid-type energy storage converter.

9. The method according to claim 1, characterized in that, Specifically, S7 includes: The virtual inertia during the fault period, the fault damping coefficient, and the active power derating coefficient in the optimal fault network control parameter combination are embedded into the corresponding control equation of the active power-frequency control loop. The reactive power support coefficient in the optimal fault network control parameter combination is embedded into the corresponding control equation of the reactive power-voltage control loop. The positive-sequence virtual impedance, negative-sequence virtual impedance, and positive-negative-sequence coupling compensation impedance parameters in the optimal fault network control parameter combination are input into the positive-negative-sequence virtual impedance control loop to correct the voltage command. The upper limit of the fixed current safety of the grid-type energy storage converter is set as the limiting boundary of the current vector limiting loop, which is used to constrain the positive and negative sequence current vector amplitudes during voltage drop faults.

10. The method according to claim 9, characterized in that, In the current vector limiting loop, a positive and negative sequence current vector circular limiting constraint is established based on the fixed safe upper limit of the current allowed by the grid-type energy storage converter: ； in, , These are the positive sequence d-axis and q-axis currents, respectively. , These are the negative sequence d-axis and q-axis currents, respectively. The upper limit of the fixed current allowed for grid-type energy storage converters; When current capacity is limited, current is allocated according to the priority of positive sequence reactive current over positive sequence active current, and positive sequence active current over negative sequence current.

Images