CN109474028B

CN109474028B - System stability optimization control method based on power grid friendly DFIG control strategy

Info

Publication number: CN109474028B
Application number: CN201811475654.4A
Authority: CN
Inventors: 张磊; 许忠义; 叶婧; 胡汉梅; 张闯; 吴罗兰; 陈俊慧; 余朋军; 徐加宝; 朱泽伟
Original assignee: China Three Gorges University CTGU
Current assignee: China Three Gorges University CTGU
Priority date: 2018-12-04
Filing date: 2018-12-04
Publication date: 2021-11-23
Anticipated expiration: 2038-12-04
Also published as: CN109474028A

Abstract

Based on the grid-friendly DFIG control strategy, the system stability optimization control method includes determining the optimal control model for the interaction effect of the traditional grid-friendly DFIG control strategy; using the spanning transient active power control model to analyze the change of DFIG active power output during system short-circuit faults ; Adopt the reactive power correction current control model with additional constraints to analyze the change of DFIG reactive power output during the system short-circuit fault; comprehensively evaluate the system stability improvement effect after optimal control. The method of the invention considers the mutual influence of the control strategies, analyzes the influence of the active power limit and recovery on the virtual inertia control effect, the virtual inertia control on the transient reactive power control effect and other factors. The strategy and the reactive power correction current control strategy with additional constraints are used to analyze the change of DFIG output during the short-circuit fault of the system, and finally the system stability improvement effect after the optimal control is comprehensively evaluated.

Description

System stability optimization control method based on power grid friendly DFIG control strategy

Technical Field

The invention relates to the field of power grid optimization control, in particular to a system stability optimization control method based on a power grid-friendly DFIG control strategy.

Background

Under the idea of power grid friendliness of a wind power plant, a DFIG (doubly Fed Induction Generator) general model comprises transient active control which can limit active power setting during a short-circuit period and an active power recovery period after the short circuit period so as to reduce load pressure of a converter and reduce over-fast active power recovery impact; and transient reactive control capable of improving reactive injection during short circuit and reducing risk of DFIG off-grid. In addition, in order to ensure the frequency stability of the system when a large active power shortage (or surplus) scene occurs in the system, virtual inertia control has also become a research hotspot in recent years. Although the design starting points of the three types of grid-friendly DFIG control strategies are different, the rotating speed (active) control time scale and the reactive control time scale in the DFIG control system are coincident with the electromechanical time scale of the system, so that when the system has a short-circuit fault, the transient active/reactive control of the DFIG can respond to the change of the terminal voltage of the DFIG, and the virtual inertia control can respond to the transient deviation of the system frequency and rapidly change the transient active and reactive output characteristics of the DFIG, thereby generating different influences on the power angle stability and the voltage drop amplitude of the system.

Although the influence of the three types of grid-friendly DFIG control strategies on the system stability is sufficiently researched, only the influence of a single control strategy on the system stability is usually concerned in the research, and the mutual influence among the control strategies is less considered, so that the grid-friendly DFIG control strategies are possibly mutually restricted in the practical application to influence the respective control effects. For example, the effect of the transient active and reactive control on the virtual inertia control effect, although the virtual inertia control has the capability of improving the stability of the power angle of the system, such an advantage is limited by the transient active and reactive control under certain conditions. As another example, the influence of the virtual inertia control on the transient reactive power control effect may affect the improvement effect of the DFIG on the voltage drop level in the fault under certain conditions.

Disclosure of Invention

In order to solve the technical problems, the invention provides a system stability optimization control method based on a power grid-friendly DFIG control strategy, which considers the mutual influence of the control strategies, analyzes factors in various aspects such as the influence of active amplitude limiting and recovery on a virtual inertia control effect, the influence of virtual inertia control on a transient reactive power control effect and the like, analyzes the DFIG output change during the short circuit fault of the system by adopting a reactive power correction current control strategy spanning transient active control strategy and additional constraint, and finally comprehensively evaluates the system stability improvement effect after optimization control.

The technical scheme adopted by the invention is as follows:

the system stability optimization control method based on the power grid friendly DFIG control strategy comprises the following steps:

step 1: aiming at the interaction influence of the traditional power grid friendly DFIG control strategy, determining an optimized control model;

step 2: analyzing the DFIG output change during the short-circuit fault of the system by adopting the optimized control model;

the step 2 comprises the following steps:

step 2.1, analyzing the DFIG active output change during the short circuit fault of the system by adopting a transient-state-spanning active control model;

step 2.2: analyzing the DFIG reactive power output change during the short circuit fault of the system by adopting a reactive power correction current control model with additional constraint;

and step 3: and comprehensively evaluating the system stability improvement effect after the optimization control.

The invention discloses a system stability optimization control method based on a power grid-friendly DFIG control strategy, which has the following beneficial effects:

1) from the perspective of power grid friendliness of the wind power plant, interaction among control strategies is fully considered, and the defect that only the influence of a single control strategy on system stability is concerned is overcome.

2) And relatively, the reactive power output of the DFIG is improved, and the risk of the DFIG being off-line is reduced.

3) The stability of the first pendulum and the reverse pendulum of the system can be improved.

4) And from a more comprehensive angle, the power angle stability of a sending end system is improved, and the voltage drop depth during the fault period is reduced.

Drawings

The invention is further illustrated by the following figures and examples.

Fig. 1 is a schematic diagram of rotor-side converter control logic.

Fig. 2 is a given graph of active during and after a fault.

Fig. 3 shows a given control diagram for the improved transient current.

Fig. 4 is an analysis diagram of the improved active control principle.

FIG. 5(a) is a schematic diagram of a time domain simulation system grid;

FIG. 5(b) is a DFIG active graph;

FIG. 5(c) is a graph of the power angle of the system;

FIG. 5(d) is a DFIG reactive plot;

FIG. 5(e) is a graph of step-up to reactive power;

FIG. 5(f) is a terminal voltage graph.

Detailed Description

the step 2 comprises the following steps:

In step 1, the conventional grid-friendly DFIG control strategy includes: transient active control, transient reactive control and virtual inertia control;

the transient active control model comprises an amplitudeValue limit i_{rd_max}，

U_measIs the DFIG terminal voltage;

the transient reactive control model comprises a transient reactive control output quantity delta i_Q：

Δi_Q＝K_v*(1-U_meas),U_meas≤0.9pu,K_v≥2 (2)

K_vIs the DFIG transient reactive gain.

When the terminal voltage of the DFIG breaks through the control dead zone, the transient reactive power control output quantity delta i_QCan be described by the formula (2). In general, to ensure the full DFIG reactive support capability during the short-circuit fault, the amplitude limiting module is switched from active priority to reactive priority, so that during the transient period i_{rd_ref1}Will also coordinate with the transient reactive control output.

Active limiting i determined by transient reactive control_{p_max}The calculation formula is as follows:

i_maxfor maximum bearable current of the converter, i_{p_max}Can send out active current to maximum when the reactive power has priority_Q0For steady state reactive current, i_{rq_ref}For q-axis active reference current, Δ i_QThe output quantity is transient reactive control output quantity.

During transient reactive control action during short-circuit fault, the steady-state reactive control integrator is locked, i_Q0Remain unchanged.

Virtual inertia control changes active reference current i_{rd_ref0}The action principle is as follows: varying torque signal T generated by a speed control system_refMeridian and L_mψ_s/L_sObtaining an active current reference i after the phase division_{rd_ref0}Wherein L is_m，ψ_s，L_sThe mutual inductance of the motor, the stator flux linkage and the self-inductance of the stator are respectively realized;

the control strategy interactions include:

(1) the influence of the magnitude relation of the two amplitude limiting effects on the transient active amplitude limiting under different conditions,

(2) the influence of active amplitude limiting and restoring on the virtual inertia control effect,

(3) the influence of the virtual inertia control on the transient reactive control effect;

a specific control strategy interaction analysis procedure is as follows,

1) analysis i_{rd_max}And i_{p_max}The magnitude relationship of the two clipping effects under different conditions:

assuming steady state operation the DFIG employs unity power factor control, i_Q0When 0.868pu ≦ U during the fault period_measWhen the content is less than or equal to 0.9pu, the following components are adopted:

when i is_{rd_max}≥i_{p_max}When there is K_v(1-U_meas) Not less than 0.2, constructor f₁(U_meas)＝0.2/(1-U_meas) When K is_vWhen the ratio is more than or equal to 2, the ratio is [0.868,0.9 ]]In the interval f₁(U_meas) The value range is [1.52,2 ]]Having K of_v≥f₁(U_meas) Is always true, i.e. K_vWhen i is more than or equal to 2_{rd_max}≥i_{p_max}Always true, so that the transient active clipping is limited by i_{p_max}And (6) determining.

Similarly, when U is less than or equal to 0.116pu_measWhen the content is less than or equal to 0.868pu, the recipe includes

When i is_{rd_max}≥i_{p_max}When there is

Constructing the right term of the inequality as a function f₂(U_meas) In the interval [0.116,0.868 ]]Its upper value range is [ -0.968,1.0987]It can be seen that K is_vEquation (6) holds constantly when i is not less than 2_{rd_max}≥i_{p_max}Constant, transient active clipping is limited by_{p_max}And (6) determining.

When U is turned_measWhen the content is less than or equal to 0.116pu, i is_{rd_max}Is 0, i.e. the active clipping is 0, in summary, since the grid-connected guidance generally requires K_vNot less than 2, so the actual active clipping i during transient_{p_lim}Comprises the following steps:

thus, in U_measAnd when the power is more than or equal to 0.116pu, the transient active maximum value is actually determined by transient reactive control.

2) Analyzing the influence of active amplitude limiting and restoring on the virtual inertia control effect:

when the frequency of the DFIG sensing system is instantaneously shifted in the short-circuit fault, the virtual inertia control output T is enabled_VIGreater than 0, such that (T)_ref-T_VI) Decrease, (T)_ref-T_VI) Corresponding active current reference and i_{p_max}The magnitude relation between the two will determine the active reference current during the transient state given i_{rd_ref2}. When U is turned_measWhen the virtual inertia control parameter K is more than or equal to 0.116pu_dWith less or less frequency shift, the active reference during the short will be i_{p_max}Determining; if the virtual inertia control parameter K_dWith a large or large frequency offset, the active current setting during the short circuit will be determined by the virtual inertia control. When U is turned_measWhen the power is less than or equal to 0.116pu, the active reference during the short-circuit fault is 0, and the active output during the short-circuit fault is 0.

After the short-circuit fault is cleared, the DFIG enters an active power recovery stage, the starting point of the DFIG recovery is related to the active current setting at the moment of clearing the short-circuit fault, and the virtual inertia control still acts at the moment when the system power angle is still in a swing stage after the short-circuit fault is considered, so that the system frequency fluctuation is caused.

Therefore, the DFIG transient active amplitude limiting control influences the virtual inertia control effect during the short circuit period, and the active recovery control influences the virtual inertia control effect after the short circuit fault, so that the DFIG transient active amplitude limiting control cannot fully respond to the system frequency deviation.

3) Analyzing the influence of the virtual inertia control on the transient reactive power control effect:

when the active power is further reduced in the short-circuit period due to the DFIG virtual inertia control, the transient reactive control output is also reduced, and finally the output apparent power of the DFIG in the short-circuit period is smaller than the rated apparent power, so that the waste of the controllable capacity of the DFIG converter in the short-circuit period is caused, and the stable supporting effect of the DFIG on the power grid in the short-circuit period is weakened.

In step 2.1, the transient-state-crossing active control model includes:

virtual inertia control output quantity T_VIDirect influencing active current reference i_{rd_ref0}Needs to be limited to i through amplitude limiting and rate-of-rise limiting_{rd_ref2}Then, the transient active power is used as the final active current reference, and the transient active power amplitude limiting and slope limiting control can be bypassed in the short-circuit period and the active power recovery period by utilizing the transient active power crossing control strategy, so that the system frequency deviation can be fully responded, and when the condition P is met_meas<P(T_ref) During the process, more active power output can be reduced during the short circuit period of the DFIG, so that the DFIG can truly respond to the system frequency, and the power angle deviation of the adjacent synchronous motor is reduced by changing the active power output of the DFIG, rather than simply carrying out active power recovery at a certain speed or curve.

In the step 2.2, the reactive power correction current control model with additional constraint comprises a reactive power correction i_{q_add}。

i_{q_add}The value of (A) takes into account two situations, when the condition P_meas<P(T_ref) And U_meas<0.9pu is given by equation (9) when both are true, otherwise it is 0,

i_{q_add}＝i_max-|i_{rd_ref2}|-|i_{rq_ref}| (9)

i_{rd_ref2}and i_{rq_ref}For the final given active and reactive current references, i_maxIs the maximum current formula (9) i_maxAnd | i_{rq_ref}I difference i active clipping i_{p_max}During the transient active control, the DFIG is actively driven (corresponding to the controlled current | i)_{rd_ref2}I) is further reduced_{q_add}Will increase and thereby change the DFIG reactive power output.

In step 3, evaluating the content includes: the optimized power angle stability and the optimized voltage drop degree.

1) And the optimized power angle stability:

compared with the first pendulum and the inverted pendulum of the system under the traditional control strategy, the power angles of the first pendulum and the inverted pendulum of the system are reduced after the additional spanning transient active control is adopted, so that the DFIG can fully respond to the deviation of the system frequency caused by the machine end when the power angle of the adjacent synchronous motor swings, and the stability of the system power angle can be effectively improved.

2) And the optimized voltage drop degree:

after reactive power correction current control of additional constraint is carried out on the basis of additional transient active control, the reactive power level of the DFIG is obviously improved in the short-circuit fault period compared with that before optimization control, the terminal voltage of the DFIG is relatively improved, and meanwhile the low-voltage ride-through capability of the DFIG can be enhanced.

Example (b):

a two-zone four-machine system was used, see fig. 5(a), with the following parameters:

1. network parameters:

in order to simulate a typical wind power delivery system in the three north area of China, in which the inertia of a receiving-end synchronous motor is far larger than that of a sending-end synchronous motor, the inertia of a synchronous motor G3 is set to be 50 times of the inertia of G1, and other parameters are the same as those of the original system. G1 is set as the PV node, and G3 is the balanced node. The synchronous motors G2 and G4 and their boost changes in the original two-zone system are removed. The line parameters and the bus voltage level are the same as the original system.

The parameter of the boost change T1 is consistent with the parameter Trf 0.69/20kV of the DFIG _2.5MW model in the DIgSIENT/Power factory Templates module, and the positive sequence reactance of the boost change T2 is 0.15 pu.

2. A double-fed asynchronous generator:

the double-fed fan model is taken from a DFIG _2.5MW model in a DIgSIENT/Power factory Templates module, but related modules are improved according to a designed control scheme, and other modules are not changed. The output of the double-fed fan is 300MW/0Mvar in a steady state, a parallel connection mode of 150 double-fed fans is adopted, and each double-fed fan outputs 2MW/0 Mvar.

3. Load and reactive power compensation device parameters:

the load L1 is 150MW/100 Mvar; the load L2 is 650MW/100 Mvar; the output of the reactive compensation device C1 is 200 Mvar; the output of the reactive compensation device C2 is 150 Mvar.

And finally, establishing a simulation model through DIgSILENT, and verifying the correctness of the influence mechanism analysis and the effectiveness of the improved control through time domain simulation.

In the short-circuit fault of the system in FIG. 1, the terminal voltage U of DFIG_measAnd phase-locked loop measuring frequency f_measAre all changed, wherein f_measWill influence the DFIG virtual inertia control output T_VI，U_measTransient active and reactive control will be affected simultaneously.

FIG. 2 Fault duration (t)₁)U_measWhen i is greater than or equal to 0.116pu_{p_max}Is the value B, then (T)_ref-T_VI) When the corresponding active reference current is a value A, the actual active current during the fault is given as a value B, namely the virtual inertia control parameter K_dWhen there is little or no frequency shift, (T)_ref-T_VI) The corresponding active reference current a will be greater than i_{p_max}(value B), so that the active reference during the fault is given by i_{p_max}And (6) determining. Similarly, when the virtual inertia control parameter K_dGreater frequency offset, greater duration of failure (T)_ref-T_VI) When the corresponding active reference current reaches the value C, the active current setting during the fault is determined by the virtual inertia control. When U is turned_measWhen the power is less than or equal to 0.116pu, the active reference during the fault is 0, and the active output during the fault is 0 and is unchanged. After the fault is cleared, the DFIG enters an active power recovery stage, and the starting point of the DFIG recovery and the fault clearingThe active current settings at the time of dividing are related, and when the active current settings at the time of fault clearing are respectively B, C, 0, the active current reference will be given according to

curves

1, 2, 3, respectively, subject to a fixed recovery rate or a fixed recovery curve after the fault. Considering that the system power angle is still in a swing stage after a fault to cause system frequency fluctuation, the virtual inertia control still acts at the moment, but T_VIIs still with T_refAnd (4) overlapping, if the obtained overlapping numerical value is higher than the numerical value corresponding to the curves 1-3 or the DFIG is recovered according to the fixed curve, the active output of the DFIG will not be influenced by virtual inertia control during active recovery.

3-4 short circuit fault, without additional improved active control, virtual inertia control signal T during fault_VIDirect and active torque reference T_refThe overlap is reflected in FIG. 4 as a decrease in the failure period (t to t) from the point X₁) Active current goes to value a. If the active current corresponding to the actual active amplitude limit during the fault period is a value C, the actual active output of the DFIG is given by the value C, and the value A determined by the virtual inertia control cannot generate an actual effect. In the active recovery stage after the fault, the DFIG recovers from the C value according to a certain rate or curve, such as curve 1, and the virtual inertia control responds to the frequency deviation of the system but from the point E, the point F and the point T_refCorresponding active current is superposed, and the obtained value is larger than the value given by the curve 1, so that the actual output of the DFIG is determined by the curve 1, and the virtual inertia control does not influence the active output of the DFIG. After the control is improved, when P is satisfied_meas<P(T_ref) When the active output of the DFIG is below the steady-state value, the virtual inertia control is directly superposed on the value C and the curve 1 (such as points Y, E ' and F '), so that the DFIG can reduce more active output (from the value C to A ') during the fault period, and can truly respond to the system frequency by superposing with the curve 1 during the active recovery period, the DFIG can change the active output per se and reduce the power angle deviation of the adjacent synchronous motor instead of purely carrying out active recovery at a certain speed or curve.

FIGS. 5(a) - (f) set the fault location at 50% between

nodes

9 and 10, fault duration 0.2s, and fault type three-phase goldAttribute ground short, DFIG transient reactive gain K_vIs 3. Four DFIG running states are set during simulation, and the working condition 1 is a virtual inertia coefficient K _d0, the recovery rate Tramp is 0.5 pu/s; operating mode 2 is K_d25 and Tramp 0.5 pu/s; working condition 3 is K_d25, Tramp is 0.5pu/s but only transient active improvement control is added; working condition 4 is K_d25, Tramp 0.5pu/s and transient active and reactive improvement control is added; observing the active curve in fig. 5(b), comparing working condition 2 with working condition 1, it can be known that after the active output of the DFIG is reduced by virtual inertia control during the fault, the active power recovery period after the fault is limited by the rate of rise, working condition 2 recovers at the same rate as working condition 1 with a lower active level, that is, a scene of a value C and a curve 2 in fig. 2 appears, and after additional improvement of transient active given control, it is equivalent to superposition of virtual inertia control output and the active curve of working condition 1, so that the active output during the transient is less, and the active power recovery period is not limited by the rate of rise, and the dynamic of the system frequency can be responded. Observing FIG. 5(d), it can be found that the transient reactive power control coefficient K of each working condition during the fault_vThe same, but different reactive power during the fault. Comparing condition 3 with condition 1 in fig. 5(e), it can be seen that during the fault, the difference between the reactive power of the low-voltage side and the reactive power of the high-voltage side of the condition 3DFIG step-up transformer is smaller, i.e. the reactive power absorbed by the condition 3DFIG step-up transformer is smaller, which is caused by the reduction of the reactive power of the DFIG during the fault, as previously seen. This phenomenon further causes a transient increase in the reactive level of the DFIG outgoing line, an increase in the DFIG terminal voltage (see fig. 5(f)), and a final decrease in the DFIG transient reactive power output.

As shown in fig. 5(c), it can be found that after the active control is additionally improved, the system power angle head pendulum and the second pendulum angle are further reduced, and compared with the working condition 2, the system head pendulum angle is reduced by about 1.4 ° and the second pendulum angle is reduced by about 6.6 ° under the working conditions 3 and 4, and the stability of the head pendulum and the second pendulum is improved. The reason for this is that as shown in fig. 5(b), compared with the working condition 2, the active power output of the DFIG is smaller under the working conditions 3 and 4 during the fault (15 s-15.2 s) and is larger during the second pendulum (15.5 s-16.4 s), the active characteristic is beneficial to the stability of the first pendulum and the second pendulum of the system, and the root of the characteristic lies in improving the active control to enable the DFIG to fully respond to the system frequency shift caused by the adjacent synchronous motor power angle swing.

As shown in fig. 5(f), compared with the working condition 1, the terminal voltage of the DFIG is higher during the fault period under the working condition 3, which is because the active power generated by the DFIG is less during the fault period under the working condition 3, the reactive power absorbed by the outgoing line is less, and the reactive power level of the outgoing line is increased for a short time, so that the terminal voltage of the DFIG is increased. After additional improved reactive power control is performed on the basis of additional improved active power control, comparing working condition 4 with working condition 3 in fig. 5(d) and 5(f), it can be known that the reactive power level of the DFIG during the fault period is further improved, the terminal voltage of the DFIG is further improved, and the low-voltage ride through capability of the DFIG is enhanced.

Claims

1. The system stability optimization control method based on the power grid friendly DFIG control strategy is characterized by comprising the following steps:

the step 2 comprises the following steps:

step 2.1, analyzing the DFIG active output change during the short circuit fault of the system by adopting an additional constrained transient active control model;

and step 3: comprehensively evaluating the system stability improvement effect after optimization control;

the transient active control model includes an amplitude limit i_{rd_max}，

U_measIs the DFIG terminal voltage;

△i_Q＝K_v*(1-U_meas),U_meas≤0.9pu,K_v≥2 (2)

K_vIs DFIG transient reactive gain;

when the terminal voltage of the DFIG breaks through the control dead zone, the transient reactive power control output quantity delta i_QCan be described by formula (2); in general, to ensure the full DFIG reactive support capability during the short-circuit fault, the amplitude limiting module is switched from active priority to reactive priority, so that during the transient period i_{rd_ref1}And also coordinate with the transient reactive control output;

i_maxfor maximum bearable current of the converter, i_{p_max}Can send out active current to maximum when the reactive power has priority_Q0For steady state reactive current, i_{rq_ref}For q-axis active reference current, Δ i_QTransient reactive control output;

during transient reactive control action during short-circuit fault, the steady-state reactive control integrator is locked, i_Q0Keeping the same;

the control strategy interactions include:

a specific control strategy interaction analysis procedure is as follows,

when i is_{rd_max}≥i_{p_max}When there is K_v(1-U_meas) Not less than 0.2, constructor f₁(U_meas)＝0.2/(1-U_meas) When K is_vWhen the ratio is more than or equal to 2, the ratio is [0.868,0.9 ]]In the interval f₁(U_meas) The value range is [1.52,2 ]]Having K of_v≥f₁(U_meas) Is always true, i.e. K_vWhen i is more than or equal to 2_{rd_max}≥i_{p_max}Always true, so that the transient active clipping is limited by i_{p_max}Determining;

When i is_{rd_max}≥i_{p_max}When there is

Constructing the right term of the inequality as a function f₂(U_meas) In the interval [0.116,0.868]Its upper value range is [ -0.968,1.0987]It can be seen that K is_vEquation (6) holds constantly when i is not less than 2_{rd_max}≥i_{p_max}Constant, transient active clipping is limited by_{p_max}Determining;

when U is turned_meas<0.116pu, when i_{rd_max}Is 0, i.e. the active clipping is 0, in summary, since the grid-connected guidance generally requires K_vNot less than 2, so the actual active clipping i during transient_{p_lim}Comprises the following steps:

thus, in U_measWhen the power is more than or equal to 0.116pu, the transient active maximum value is actually determined by transient reactive control;

when the frequency of the DFIG sensing system is instantaneously shifted in the short-circuit fault, the virtual inertia control output T is enabled_VIGreater than 0, such that (T)_ref-T_VI) Decrease, (T)_ref-T_VI) Corresponding active current reference and i_{p_max}The magnitude relation between the two will determine the active reference current during the transient state given i_{rd_ref2}；

When U is turned_measWhen the virtual inertia control parameter K is more than or equal to 0.116pu_dWith less or less frequency shift, the active reference during the short will be i_{p_max}Determining; if the virtual inertia control parameter K_dLarger or larger frequency offset, the active current given during the short circuit will be determined by the virtual inertia control; when U is turned_meas<When the voltage is 0.116pu, the active reference during the short-circuit fault is 0, and the active output during the short-circuit fault is 0;

after the short-circuit fault is cleared, the DFIG enters an active power recovery stage, the starting point of the DFIG recovery is related to the active current setting at the moment of clearing the short-circuit fault, and the virtual inertia control still acts at the moment when the system power angle is still in a swing stage after the short-circuit fault is considered to cause system frequency fluctuation;

therefore, the DFIG transient active amplitude limiting control influences the virtual inertia control effect during the short circuit period, and the active recovery control influences the virtual inertia control effect after the short circuit fault, so that the DFIG transient active amplitude limiting control cannot fully respond to the frequency deviation of the system;

when active power is further reduced in the short-circuit period due to the DFIG virtual inertia control, transient reactive control output is also reduced, and finally the output apparent power of the DFIG in the short-circuit period is smaller than the rated apparent power, so that the waste of the controllable capacity of a DFIG converter in the short-circuit period is caused, and the stable supporting effect of the DFIG on a power grid in the short-circuit period is weakened;

in step 2.1, the transient active control model across additional constraints includes:

virtual inertia control output quantity T_VIDirect influencing active current reference i_{rd_ref0}Needs to be limited to i through amplitude limiting and rate-of-rise limiting_{rd_ref2}Then, the transient active power is used as the final active current reference, and the transient active power amplitude limiting and slope limiting control can be bypassed in the short-circuit period and the active power recovery period by utilizing the transient active power crossing control strategy, so that the system frequency deviation can be fully responded, and when the condition P is met_meas<P(T_ref) During the process, more active power output can be reduced during the short circuit period of the DFIG, so that the DFIG can truly respond to the system frequency, and the power angle deviation of the adjacent synchronous motor is reduced by changing the active power output of the DFIG, rather than performing active power recovery at a certain speed or curve;

in the step 2.2, the reactive power correction current control model with additional constraint comprises a reactive power correction i_{q_add}；

i_{q_add}＝i_max-|i_{rd_ref2}|-|i_{rq_ref}| (9)

i_{rd_ref2}and i_{rq_ref}For the final given active and reactive current references, i_maxIs the maximum current;

in the formula (9): i_maxAnd | i_{rq_ref}I difference i active clipping i_{p_max}When the DFIG active power output during the short circuit is further reduced by crossing the transient active control, i_{q_add}Will increase and thereby increase the DFIG reactive power output.

2. The optimal control method for the system stability under the power grid-friendly DFIG control strategy according to claim 1, wherein the optimal control method comprises the following steps: in step 3, evaluating the content includes: the optimized power angle stability and the optimized voltage drop degree;

1) and the optimized power angle stability:

compared with the first pendulum and the inverted pendulum of the system under the traditional control strategy, after the additional spanning transient active control is adopted, the power angles of the first pendulum and the inverted pendulum of the system are reduced, so that the DFIG can fully respond to the offset of the system frequency caused by the machine end when the power angle of the adjacent synchronous motor swings, and the stability of the system power angle can be effectively improved;

2) and the optimized voltage drop degree: