Chain path analysis method and system in doubly-fed wind farm considering strategy switching

Abstract

The present invention belongs to the technical field of wind power generation, and specifically discloses a chain path analysis method and system in a doubly fed wind farm considering strategy switching, the method comprising: under a conventional operation strategy, calculating the probability of each DFIG in the wind farm being disconnected from the grid due to a crowbar protection action; under a low voltage ride-through operation strategy, calculating the terminal voltage of each DFIG, and judging the next DFIG to be disconnected from the grid and the time of disconnection; after the topology structure changes, re-judging the next DFIG to be disconnected from the grid and the time of disconnection until all DFIGs are disconnected from the grid; considering the switching from the conventional operation strategy to the low voltage ride-through operation strategy, establishing a deductive relationship between the state probability distribution of two consecutive disconnection events based on a transfer matrix; based on the action probability of each DFIG in the wind farm being disconnected from the grid due to a crowbar protection action, solving the initial state probability distribution and the transfer matrix, and then obtaining the state probability distribution of each disconnection event; the present invention can reflect the chain disconnection path analysis of the wind speed difference at each position in the wind farm.

Description

Technical Field

[0001] The present invention relates to the technical field of wind power generation, and in particular to a chain path analysis method and system within a double-fed wind farm considering strategy switching. Background Art

[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] Wind power generation technology represented by Doubly-Fed Induction Generator (DFIG) is becoming more mature, and the proportion of installed capacity is increasing year by year. The high penetration rate of wind power generation makes the dynamic characteristics of the power system, including cascading failures, more complex.

[0004] Cascading failures are a mixed process in which segmented continuous dynamics and discrete events occur alternately and sequentially. In existing studies on cascading failures of renewable energy power systems, the possible off-grid events of multiple wind turbines in a wind farm are usually regarded as one-time events that occur simultaneously, thereby ignoring the dynamic differences and mutual influences of the units in the field and the resulting cascading off-grid processes in the wind farm. On the one hand, according to the multi-time scale singular perturbation theory, when this local cascading process is much faster than the transient state of the power system, the error caused by ignoring the former is small. However, with the continuous increase of power electronic equipment in power systems, some research scenarios need to explore the system characteristics dominated by transients at shorter time scales, and ignoring the local dynamics in renewable energy sites will be unacceptable. On the other hand, with the flexibility of power electronic control, DFIG usually configures a variety of disturbed operating strategies in addition to conventional operating strategies, such as the strategy of outputting reactive power to support the grid voltage during low voltage ride-through. When the DFIG is disturbed and the trigger time or state conditions are met, the operation strategy will switch. Mathematically, the strategy switch means the change of the dynamic equation, which can significantly affect the DFIG state motion trajectory and the state coupling relationship between multiple DFIGs, thereby changing the development path of chain events in the wind farm.

[0005] Therefore, based on the above two analyses, local chain events within a doubly fed wind farm should be given full attention, and considering the new characteristics that may be brought about by strategy switching, the impact of strategy switching on the chain path within the wind farm during low voltage ride-through is also worthy of in-depth discussion.

[0006] In terms of system-level cascading failure research, most existing research work does not consider the dynamic characteristics of wind turbines, or even if it is considered, it is assumed that the actions of all units in the entire wind farm occur simultaneously and consistently, ignoring the local dynamic differences within the field. In terms of station-level cascading failure research, existing research focuses on the impact of crowbar protection and reactive power compensation equipment on wind turbine cascading disconnection. There is a lack of detailed analysis of the impact of reactive power and collector line impedance issued according to national standards on cascading disconnection during wind turbine low-voltage ride-through. In terms of switching system dynamic analysis, existing research mainly focuses on how switching caused by the limiting link affects system stability, while there is little discussion on the impact of control strategy switching on system dynamics, especially on cascading failures within wind farms. Summary of the invention

[0007] In order to solve the above problems, the present invention proposes a chain path analysis method and system within a doubly fed wind farm considering strategy switching, which can evaluate the impact of switching from a conventional operation strategy with crowbar protection to a low voltage ride-through operation strategy with low voltage protection on the chain path within the wind farm, and deduce the chain disconnection process of wind turbines considering strategy switching, so as to formulate corresponding measures in advance to block chain failures within the wind farm.

[0008] In some embodiments, the following technical solutions are adopted:

[0009] A chain path analysis method in a doubly-fed wind farm considering strategy switching includes:

[0010] Under the conventional operation strategy, the probability of each DFIG in the wind farm being disconnected from the grid due to the crowbar protection action is calculated;

[0011] Under the low voltage ride-through operation strategy, the terminal voltage of each DFIG is calculated, and the next DFIG to be disconnected and the time of disconnection are determined; after the topology structure changes due to the unit being disconnected, the impedance matrix is ​​corrected, and the next DFIG to be disconnected and the time of disconnection are re-determined until all DFIGs are disconnected;

[0012] Considering the switch from the conventional operation strategy to the low voltage ride-through operation strategy, the probability distribution V of two consecutive off-grid events is established based on the transfer matrix T. m and V m+1 Based on the probability of each DFIG in the wind farm being disconnected from the grid due to the crowbar protection action, the initial state probability distribution V1 and the transfer matrix T are solved, and then the state probability distribution of each disconnection event is obtained.

[0013] As a further solution, under the conventional operation strategy, the probability of each DFIG in the wind farm being disconnected from the grid due to the crowbar protection action is calculated, specifically:

[0014]

[0015] Among them, P m1 , P m2 They are the minimum power and maximum power of DFIG when operating at MPPT;

[0016] P c is the critical active power of DFIG for crowbar protection action, specifically:

[0017]

[0018] f P (P) is the probability density function of the active power output when DFIG operates in MPPT, specifically:

[0019]

[0020] in, ρ is the air density; S w is the windward swept area of ​​the wind turbine blades; C pmax is the maximum wind energy utilization coefficient; opt is the optimal tip speed ratio; n p is the number of DFIG pole pairs; N is the gearbox speed ratio; P sc is the critical stator active power when the crowbar protection is activated; R w is the radius of the wind wheel; ω1 is the synchronous angular frequency; k is the shape parameter; c is the scale parameter; P is the DFIG active power.

[0021] As a further solution, under the low voltage ride-through operation strategy, the terminal voltage of each DFIG is calculated as follows:

[0022] U=U p +ZI;

[0023] Among them, U and I are matrices composed of the terminal voltage and output reactive current of each DFIG respectively; U p is the matrix composed of PCC point voltages; Z is the impedance matrix formed by the branch addition method according to the wind farm topology. When the topology changes due to the unit being disconnected from the grid, the impedance matrix Z needs to be recalculated.

[0024] As a further solution, the next off-grid DFIG and off-grid time are determined according to the relationship between the DFIG off-grid time t and the terminal voltage, specifically:

[0025] The relationship between the DFIG off-grid moment and the terminal voltage is described as: t = 1.96U t +0.23(0.2≤U t ≤0.9);

[0026] Among them, U t is the per unit voltage at the wind turbine terminal.

[0027] As a further solution, under the low voltage ride-through operation strategy, the chain disconnection of multiple DFIGs in the field is manifested as: sequential disconnection from the point close to the PCC point to the point far away from the PCC point, and the disconnection speed is getting faster and faster, showing an accelerated development.

[0028] As a further solution, the probability distribution V of two consecutive off-line events is established based on the transfer matrix T. m and V m+1 The deductive relationship is as follows:

[0029] TV m =V m+1 ;

[0030] Among them, V m 、V m+1 are the state probability distributions of the mth and m+1th events respectively;

[0031] Assume that there are n DFIGs, and define the set of optional states E when each event occurs as: E = {S1,…,S k ,…,S i ,…,S n ,S n+1}; where S1,…,S n Respectively represent the first DFIG to the nth DFIG off-grid, S n+1 Indicates that all DFIGs are off-grid; the state is divided into absorption state and transfer state; when in the absorption state, the transfer probability to other states is 0, and it will stay in this state; when in the transfer state, it will transfer to other states according to the transfer probability when the next event occurs; state S1,…,S n is the transition state, state S n+1 In absorption state.

[0032] As a further solution, based on the probability of each DFIG in the wind farm being disconnected from the grid due to the crowbar protection action, the initial state probability distribution V1 is solved, which is:

[0033] The direction close to the PCC point is described as the front side, and the direction away from the PCC point is described as the back side;

[0034] When the initial event is in the transition state S1,…,S n When the state S k The probability P k (1) For example, after the strategy is switched, the DFIG that is closer to the PCC point will be disconnected earlier, so P k (1)This is equivalent to the probability that all DFIGs in front of the kth DFIG trigger the crowbar protection, while the kth DFIG does not trigger the crowbar protection, that is, q k is the probability of the kth DFIG being disconnected from the grid due to the crowbar protection action;

[0035] When the initial event is in the absorbing state S n+1 When , it means that all DFIGs trigger the crowbar protection to disconnect from the grid, that is

[0036] As a further solution, based on the probability of each DFIG in the wind farm being disconnected from the grid due to the crowbar protection action, the transfer matrix T is solved, which is:

[0037] To solve the transition state S k , k≤n transfer to the probability of various states in the next event as an example:

[0038] Since the transfer state will not be transferred to the state of the next event again, and after the strategy switching, the DFIG is disconnected from the grid in sequence from the electrical distance to the PCC point from near to far, the state S k It is impossible to transfer to the state S1,…,S of the next event k , that is, P 1k ,…,P kk All are 0;

[0039] By state S k Transfer to state S in the next event i Probability P ik It is equivalent to the probability that all other DFIGs between the kth DFIG and the ith DFIG trigger the crowbar protection and the ith DFIG does not trigger the crowbar protection, that is, i>k;

[0040] By state S k Transfer to the next event absorbing state S n+1 The probability P n+1k It is equivalent to the probability that all DFIGs behind the kth DFIG trigger the crowbar protection, that is,

[0041] When the nth DFIG at the end of the collector line is disconnected from the grid, all DFIGs in the wind farm are also disconnected from the grid, and the state must be changed from S n Transfer to state S in the next event n+1 , that is, P n+1n =1;

[0042] When the event is in the absorbing state S n+1 When P n+1,n+1 =1.

[0043] In other embodiments, the following technical solutions are adopted:

[0044] A chain path analysis system in a doubly-fed wind farm considering strategy switching, comprising:

[0045] The DFIG off-grid analysis module in the conventional operation strategy stage is used to calculate the probability of each DFIG in the wind farm being off-grid due to the crowbar protection action under the conventional operation strategy;

[0046] The DFIG off-grid analysis module in the low voltage ride-through operation strategy stage is used to calculate the terminal voltage of each DFIG under the low voltage ride-through operation strategy, and determine the next off-grid DFIG and the off-grid time; after the topology structure changes due to the unit off-grid, the impedance matrix is ​​corrected, and the next off-grid DFIG and the off-grid time are re-determined until all DFIGs are off-grid;

[0047] The off-grid analysis module in the DFIG operation strategy switching phase is used to consider the switch from the conventional operation strategy to the low voltage ride-through operation strategy. Based on the transfer matrix T, the probability distribution V of the state of two consecutive off-grid events is established. m and V m+1 Based on the probability of each DFIG in the wind farm being disconnected from the grid due to the crowbar protection action, the initial state probability distribution V1 and the transfer matrix T are solved, and then the state probability distribution of each disconnection event is obtained.

[0048] In other embodiments, the following technical solutions are adopted:

[0049] A terminal device comprises a processor and a memory, wherein the processor is used to implement instructions; the memory is used to store a plurality of instructions, wherein the instructions are suitable for being loaded by the processor and executing the above-mentioned chain path analysis method within a doubly-fed wind farm considering strategy switching.

[0050] Compared with the prior art, the present invention has the following beneficial effects:

[0051] (1) The present invention analyzes in detail the evolution law of cascading faults in a doubly fed wind farm during conventional operation strategy, low voltage ride-through operation strategy and strategy switching, analyzes the randomness of DFIG disconnection due to crowbar protection action in a doubly fed wind farm under conventional operation strategy, and the cascading sequentiality and acceleration characteristics of multiple DFIGs disconnecting due to low voltage protection action in a doubly fed wind farm under low voltage ride-through operation strategy; by analyzing the way and reason why strategy switching changes the DFIG cascading disconnection path in a doubly fed wind farm, the DFIG cascading disconnection process considering strategy switching is deduced, so as to formulate corresponding measures in advance to block cascading faults in the wind farm.

[0052] (2) The method of the present invention takes into account the impact of DFIG strategy switching on the chain disconnection path, and can realize the chain disconnection path analysis that reflects the wind speed differences at various locations in the wind farm.

[0053] Other features and advantages of additional aspects of the present invention will be given in part in the following description, and in part will become obvious from the following description, or will be learned through the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 It is a flow chart of a chain path analysis method in a doubly-fed wind farm considering strategy switching in an embodiment of the present invention;

[0055] Figure 2 Schematic diagram of the grid-connected time length regulation of low voltage ride-through of a wind turbine generator system in an embodiment of the present invention;

[0056] Figure 3 It is a schematic diagram of multiple DFIGs accelerating interlocking off-grid during low voltage ride-through in an embodiment of the present invention;

[0057] Figure 4 A schematic diagram of strategy switching and off-grid event transition in a doubly-fed wind farm according to an embodiment of the present invention;

[0058] Figure 5 A double-fed wind farm model used for case verification in an embodiment of the present invention;

[0059] Figure 6 Schematic diagram of rotor current of each DFIG in a wind farm after voltage drop in an embodiment of the present invention;

[0060] Figure 7 It is a schematic diagram of the change of the terminal voltage of each DFIG and the total reactive power output of the wind farm during the low voltage ride-through period in an embodiment of the present invention;

[0061] Figure 8 It is a schematic diagram comparing the probability of crowbar protection action of each DFIG in the embodiment of the present invention and the frequency of event occurrence in the sampling simulation;

[0062] Fig. 9 Schematic diagram comparing the theoretical calculated state probability distribution of each event in the embodiment of the present invention and the state event occurrence frequency distribution obtained by simulation. DETAILED DESCRIPTION

[0063] It should be noted that the following detailed descriptions are illustrative and are intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used in the present invention have the same meanings as those commonly understood by those skilled in the art to which the present application belongs.

[0064] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0065] Embodiment 1

[0066] In one or more embodiments, a chain path analysis method for a doubly fed wind farm considering strategy switching is disclosed, combined with Figure 1 , specifically including the following process:

[0067] (1) Under the conventional operation strategy, the probability of each DFIG in the wind farm being disconnected from the grid due to the crowbar protection action is calculated.

[0068] Specifically, under normal grid-connected conditions, DFIG works with the conventional operation strategy to ensure MPPT, usually operating in a unity power factor mode with priority output of active power, and configured with crowbar protection. If the rotor current exceeds the protection setting value, the crowbar protection is activated and the DFIG is disconnected from the grid within about 30ms.

[0069] After the grid voltage drops, the DFIG equivalent circuit is divided into the stator terminal voltage U s0 The normal operation equivalent circuit and the stator terminal voltage are -xU s0 The fault equivalent circuit is: rf Described as:

[0070] I rf =I r0 +ΔI r (1)

[0071] In the formula, I r0 is the rotor current in the normal operation equivalent circuit; ΔI r is the rotor current in the fault equivalent circuit.

[0072] In the normal operation equivalent circuit, the rotor current is described in the dq rotating coordinate system as:

[0073]

[0074] Where, L s is the stator inductance; L m is mutual inductance; P s0 is the stator power before the fault; U s0 is the stator voltage vector amplitude before the fault; ω1 is the synchronous angular frequency; i rd、i rq They are the d-axis and q-axis components of the rotor current respectively.

[0075] The voltage and flux equations in the fault equivalent circuit are described as:

[0076]

[0077] In the formula, R s , R r are stator and rotor resistance respectively; ΔI s , ΔI r are the stator and rotor current vectors respectively; Δψ s , Δψ r are the stator and rotor flux vectors respectively; s is the slip rate; L r is the rotor equivalent inductance; t is the time.

[0078] ΔI is obtained from formula (3): r for:

[0079]

[0080] In the formula,

[0081] Substituting equations (2) and (4) into equation (1), we can obtain the rotor current amplitude after the fault: rfmax for:

[0082]

[0083] In the formula, u sd0 、u sq0 They are the d-axis and q-axis components of the stator terminal voltage before the fault.

[0084] Assume the rotor crowbar protection action constant is I rc , let I in formula (5) rfmax =I rc , the critical stator active power P when the crowbar protection is activated sc Described as:

[0085]

[0086] Ignoring mechanical loss and stator copper loss, the basic formula for DFIG operation at MPPT is described as:

[0087]

[0088] In the formula, P, P s , P m are DFIG active power, stator power and mechanical power respectively; ρ is air density; S wis the windward swept area of ​​the wind turbine blades; C pmax is the maximum wind energy utilization coefficient; v is the wind speed; R w is the radius of the wind wheel; ω r is the rotor angular frequency; λ opt is the optimal tip speed ratio; n p is the number of DFIG pole pairs; N is the gearbox speed ratio.

[0089] Combining equations (6) and (7), the critical active power P of the DFIG in the crowbar protection action is c Described as:

[0090]

[0091] In the formula,

[0092] The wind speed approximately follows the Weibull distribution, and its probability density function f v (v) Described as:

[0093]

[0094] Where k is the shape parameter and c is the scale parameter.

[0095] According to equations (8) and (9), the probability density function f of the active power output when DFIG operates in MPPT is P (P) is described as:

[0096]

[0097] According to equations (8) and (10), the crowbar protection action probability q when DFIG operates in MPPT is described as:

[0098]

[0099] Where P m1 , P m2 They are the minimum and maximum power of DFIG when operating in MPPT.

[0100] The independent probability V0 of multiple DFIGs in a wind farm being disconnected from the grid due to crowbar protection action is described as:

[0101] V0=[q1 … q k … q i … q n ] T (12)

[0102] In the formula, q k is the probability of the kth DFIG being disconnected from the grid due to the crowbar protection action.

[0103] (2) Under the low voltage ride-through operation strategy, the terminal voltage of each DFIG is calculated, and the next DFIG to be disconnected and the time of disconnection are determined based on the relationship between the DFIG disconnection time and the terminal voltage. After the topology structure changes due to the unit disconnection, the impedance matrix is ​​corrected, and the next DFIG to be disconnected and the time of disconnection are re-determined until all DFIGs are disconnected.

[0104] Specifically, when the DFIG detects that the terminal voltage drops below 0.9pu, it will switch to the low voltage ride-through operation strategy. Figure 2 The national standard requires the configuration of low voltage protection. Figure 2 When the voltage is above the medium voltage contour line, the wind turbine remains connected to the grid and generates reactive current according to equation (13) to support the grid voltage. When the voltage is below the voltage contour line, the configured low voltage protection is triggered and the wind turbine is disconnected from the grid.

[0105] ΔI t =K1(0.9-U t ),(0.2≤U t ≤0.9) (13)

[0106] In the formula, ΔI t is the output reactive current increment; K1 is the reactive current proportional coefficient; U t is the per unit voltage at the wind turbine terminal.

[0107] Consider that severe external disturbances cause the voltages of multiple DFIG units in the field to drop to the range of 0.2pu to 0.9pu, and the external disturbance lasts for a long time (for example, the external protection fails to operate). Figure 2 The middle oblique line voltage contour line describes the relationship between the DFIG off-grid moment and the generator terminal voltage as:

[0108] t=1.96U t +0.23(0.2≤U t ≤0.9) (14)

[0109] Due to the existence of the collector line impedance and the reactive power generated by the DFIG, the terminal voltage distribution of each DFIG is uneven. The voltage distribution model is described as:

[0110] U=U p +ZI (15)

[0111] Where U and I are matrices composed of the terminal voltage and output reactive current of each DFIG respectively; U p is the matrix composed of PCC point voltage; Z is the impedance matrix formed by the branch addition method according to the wind farm topology. When the topology changes due to the unit being disconnected from the grid, the impedance matrix needs to be recalculated. The specific expressions of each matrix are:

[0112]

[0113] In the formula, u k and i k They represent the terminal voltage and output current of the kth DFIG, u pcc represents the PCC point voltage, z i Indicates the collector line impedance between adjacent units.

[0114] The terminal voltage of each DFIG is calculated by equation (15), and the next disconnected DFIG and the time of disconnection are determined according to equation (14). Considering the change of field topology and the disappearance of output reactive current caused by the disconnection of this DFIG, after correcting the impedance matrix, the next disconnected DFIG is determined again by equations (15) and (14), thus forming a loop iteration until all DFIGs are exited, and the complete process of low voltage ride-through chain disconnection of wind farms can be obtained.

[0115] The chain disconnection of multiple DFIGs in the field is characterized by sequential disconnection from the PCC point to the PCC point, and the disconnection speed is getting faster and faster, showing an accelerated development. The mechanism is that due to the existence of the collector line impedance and the flow of reactive power to the PCC point, the closer the DFIG is to the PCC point, the lower the voltage is; as the chain disconnection process develops, when the terminal voltage of a DFIG is lower than Figure 2 When the voltage contour line is reached, the DFIG is immediately disconnected from the grid due to low voltage protection action, and the reactive current it generates also disappears. Other DFIGs lose the reactive power support of this disconnected DFIG, and the voltage at the machine end further drops. The disconnection time is advanced, and the chain disconnection develops at an accelerated rate. Figure 3 shown.

[0116] (3) Considering the switch from the conventional operation strategy to the low voltage ride-through operation strategy, the probability distribution V of two consecutive off-grid events is established based on the transfer matrix T. m and V m+1 Based on the probability of each DFIG in the wind farm being disconnected from the grid due to the crowbar protection action, the initial state probability distribution V1 and the transfer matrix T are solved, and then the state probability distribution of each disconnection event is obtained.

[0117] Specifically, after a short period of time when the low voltage disturbance occurs, the DFIG detects a voltage drop at the generator end, and the conventional operation strategy switches to the low voltage ride-through operation strategy, such as Figure 4 As shown in the figure, the randomness of DFIG triggering crowbar protection disconnection results in that the order of each DFIG triggering low voltage protection during the low voltage ride-through phase and causing accelerated chain disconnection is no longer fixed after the strategy is switched.

[0118] The medium voltage collection line in the double-fed wind farm usually adopts a chain structure. Figure 3 There are n DFIGs in the field collector circuit shown. Based on this structure, a DFIG chain disconnection deduction model in the low voltage ride-through operation stage is established based on the Markov chain. The optional state set E (sample space) when each event occurs in the Markov chain deduction model is defined as:

[0119] E={S1,…,S k ,…,S i ,…,S n ,S n+1} (17)

[0120] Where S1,…,S n Respectively represent the first DFIG to the nth DFIG off-grid, S n+1 Indicates that all DFIGs are off-grid. The state can be divided into absorption state and transfer state. When in the absorption state, the transfer probability to other states is 0, and it will stay in this state; when in the transfer state, it will transfer to other states according to the transfer probability when the next event occurs. For the process of multiple DFIGs off-grid in a double-fed wind farm under severe external disturbance, states S1,…,S n is the transition state, state S n+1 In absorption state.

[0121] The relationship between the probability distribution V of two consecutive off-line events based on the transfer matrix T is described as follows:

[0122] TV m =V m+1 (18)

[0123] Where:

[0124]

[0125] P ik is the state S in the previous event k Transfer to the subsequent event state S i The probability of V m 、V m+1 are the state probability distributions of the mth and m+1th events respectively; P k (m) , P k (m+1) The mth and m+1th events are in state S respectively k probability.

[0126] Based on the probability q of each DFIG crowbar protection action, the initial state probability distribution V1 and the transfer matrix T are solved.

[0127] (a) Solution for the state probability distribution V1 of the initial event

[0128] For the convenience of discussion, the direction close to the PCC point is described as the front side, and the direction away from the PCC point is described as the back side.

[0129] When the initial event is in the transition state S1,…,S n When the state S k The probability P k (1) For example, after the strategy is switched, the DFIG that is closer to the PCC point will be disconnected earlier, so P k (1) This is equivalent to the probability that all DFIGs in front of the kth DFIG trigger the crowbar protection, while the kth DFIG does not trigger the crowbar protection, that is,

[0130] When the initial event is in the absorbing state S n+1 When , it means that all DFIGs trigger the crowbar protection to disconnect from the grid, that is

[0131] In summary, the state probability distribution V1 of the initial event is described as:

[0132]

[0133] (b) Solving the transfer matrix T

[0134] To solve the transition state S k (where k≤n) is used as an example to calculate the probability of transferring to various states in the next event.

[0135] ① Since the transfer state will not be transferred to the state of the next event again, and after the strategy switching, the DFIG is disconnected from the grid in sequence from the electrical distance to the PCC point from near to far, the state S k It is impossible to transfer to the state S1,…,S of the next event k , that is, P 1k ,…,P kk Both are 0.

[0136] ② From state S k Transfer to state S in the next event i (where i>k) the probability P ik It is equivalent to the probability that all other DFIGs between the kth DFIG and the ith DFIG trigger the crowbar protection and the kth DFIG does not trigger the crowbar protection, that is,

[0137] ③ From state S k Transfer to the next event absorbing state S n+1 The probability P n+1,kIt is equivalent to the probability that all DFIGs behind the kth DFIG trigger the crowbar protection, that is,

[0138] In particular, when the last (nth) DFIG at the end of the collector line is disconnected from the grid, all DFIGs in the wind farm are also disconnected from the grid, and the state S n Transfer to state S in the next event n+1 , that is, P n+1,n = 1. When the event is in the absorbing state S n+1 When it is, it will stay in this state, that is, P n+1,n+1 =1.

[0139] The transfer matrix T is described as:

[0140]

[0141] The following is a simulation analysis of the method of this embodiment:

[0142] (1) Example 1: Verification of accelerated interlocking disconnection of multiple DFIGs in a wind farm

[0143] Based on an actual wind farm, a double-fed wind farm model consisting of 11 DFIGs was built. Figure 5 The rated power of a single DFIG is 1.50MW, the rated voltage is 0.69kV, and other main parameters are listed in Table 1. Each DFIG in the wind farm adopts a separate control strategy, and is configured with a conventional operation strategy with crowbar protection, as well as low voltage protection and low voltage ride-through operation strategies according to regulations.

[0144] In this example, the DFIG output is 1MW, and the crowbar protection action threshold is 2.32pu. A three-phase grounding short circuit occurred on the 110kV busbar on the high-voltage side of the wind farm main transformer at 0s, and the voltage dropped to 0.46pu. Under the influence of this disturbance, the rotor currents of the 11 DFIGs in the field are as follows: Figure 6 As shown in the figure, all of them are lower than the crowbar action value, and all DFIG crowbar protections are not activated. All DFIGs then switched to low voltage ride-through operation, generating reactive power to support the grid voltage.

[0145] Table 1 Main parameters of a single DFIG

[0146]

[0147]

[0148] Figure 7The figure shows the time variation of the voltage at each DFIG terminal (solid line, corresponding to the left ordinate) and the total reactive power output of the wind farm (dashed line, corresponding to the right ordinate) during the low voltage ride-through process. Each significant drop in the total reactive power output means that a DFIG in the farm is disconnected from the grid, the voltage at the terminal of the DFIG drops to zero, and the reactive power output also disappears. For example, Figure 7 The reactive power output declines as a DFIG is disconnected from the grid, which will also affect the terminal voltage of other DFIGs. Therefore, at the above moments, the terminal voltage of other DFIGs on the grid will also drop significantly. Figure 7 It can also be seen that due to the existence of the impedance of the collector line in the field, the voltage at the end of each DFIG is not the same. The closer the DFIG is to the PCC point, the lower the voltage at the end, and the earlier it is disconnected from the grid. The previous DFIG disconnection event can significantly accelerate the subsequent DFIG disconnection process. When DFIG1 was disconnected at 1.319s, the reactive power it generated disappeared, and the reactive power output of the wind farm dropped from 6.20MVar to 5.65MVar. The voltage at the end of DFIG2 dropped from 0.559pu to 0.556pu, and its disconnection time was also advanced from 1.326s to 1.320s, 6ms earlier. Overall, the interval between two adjacent disconnection events in the wind farm also gradually shortened from 5.98ms at the beginning of the DFIG1 and DFIG2 disconnection interval to 0.05ms at the end of the DFIG10 and DFIG11 disconnection interval.

[0149] The results of this simulation case confirm the phenomenon of multiple DFIGs accelerating chain disconnection from the grid that may be caused by the low voltage ride-through operation strategy in the wind farm analyzed in the previous article.

[0150] (2) Example 2: Verification of the simulation model of multiple DFIGs disconnecting from the grid in a field considering strategy switching

[0151] Considering that the wind speed at the location of each DFIG in the wind farm may be different, it is assumed that the shape parameter k or scale parameter c in the Weibull wind speed probability density function corresponding to each DFIG is different, as shown in Table 2. The proposed deduction model is verified based on the Monte Carlo method. In order to show the process of the Monte Carlo method gradually approaching the probability distribution, 10, 100, and 1000 sampling simulations were performed respectively. For the convenience of description, this example only retains DFIG1 to DFIG5 of Example 1.

[0152] Table 2 Shape parameter k and scale parameter c of wind speed probability density function corresponding to each DFIG

[0153]

[0154] Figure 8The following is a comparison chart of the calculated probability and sampling simulation of the crowbar protection action of each DFIG. Taking DFIG2 as an example, the deviations between the frequency of crowbar protection action events obtained by 10, 100 and 1000 sampling simulations and the theoretical calculated probability are 0.235, 0.015 and 0.012 respectively, indicating that with the increase in the number of sampling simulations, the frequency of DFIG2 crowbar protection action events gradually exceeds the theoretical calculated probability. It also verifies the probability of crowbar protection action obtained in this embodiment.

[0155] Since each DFIG in the field is randomly disconnected from the grid with a certain probability under the protection of the crowbar, when the strategy is switched to low voltage ride-through, the DFIG sequential accelerated disconnection will also show certain random characteristics. The probability distribution of each disconnection event in the low voltage ride-through strategy stage of the double-fed wind farm is obtained according to the Markov deduction, and the frequency of events obtained by sampling simulation is compared. Fig. 9 As shown in (a)-(f) in the figure; wherein, (a) represents the calculated state probability distribution value, 10, 100 and 1000 sampling simulation value results of the 1st to 5th DFIG being disconnected from the grid under the first off-grid event; similarly, (b)-(f) represent the calculated state probability distribution value, 10, 100 and 1000 sampling simulation value results of the 1st to 5th DFIG being disconnected from the grid under the second, third, fourth, fifth and sixth off-grid events, respectively.

[0156] by Fig. 9 Taking the second event in state S2 in (b) as an example, that is, DFIG2 is the second DFIG to go offline, the deviations between the occurrence frequencies obtained by 10, 100 and 1000 simulations and the theoretical calculated probability are 0.130, 0.010 and 0.006 respectively, indicating that with the increase of the number of sampling simulations, the frequency of the second event in state S2 approaches the theoretical calculated probability.

[0157] The simulation analysis verifies the correctness of the DFIG interlocking off-grid simulation model in the double-fed wind farm proposed in this embodiment.

[0158] Embodiment 2

[0159] In one or more embodiments, a chain path analysis system in a doubly-fed wind farm considering strategy switching is disclosed, comprising:

[0160] The DFIG off-grid analysis module in the conventional operation strategy stage is used to calculate the probability of each DFIG in the wind farm being off-grid due to the crowbar protection action under the conventional operation strategy;

[0161] The DFIG off-grid analysis module in the low voltage ride-through operation strategy stage is used to calculate the terminal voltage of each DFIG under the low voltage ride-through operation strategy, and determine the next off-grid DFIG and the off-grid time; after the topology structure changes due to the unit off-grid, the impedance matrix is ​​corrected, and the next off-grid DFIG and the off-grid time are re-determined until all DFIGs are off-grid;

[0162] The off-grid analysis module in the DFIG operation strategy switching phase is used to consider the switch from the conventional operation strategy to the low voltage ride-through operation strategy. Based on the transfer matrix T, the probability distribution V of the state of two consecutive off-grid events is established. m and V m+1 Based on the probability of each DFIG in the wind farm being disconnected from the grid due to the crowbar protection action, the initial state probability distribution V1 and the transfer matrix T are solved, and then the state probability distribution of each disconnection event is obtained.

[0163] It should be noted that the specific implementation method of the above modules is the same as that in Example 1 and will not be described in detail.

[0164] Embodiment 3

[0165] In one or more embodiments, a terminal device is disclosed, including a server, the server including a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the chain path analysis method in a doubly-fed wind farm considering strategy switching in Embodiment 1 when executing the program. For the sake of brevity, it will not be described in detail here.

[0166] Although the above describes the specific implementation mode of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art on the basis of the technical solution of the present invention without creative work are still within the scope of protection of the present invention.

Claims

1. A chain path analysis method within a doubly-fed wind farm considering strategy switching, characterized in that: include: Under the conventional operation strategy, the probability of each DFIG in the wind farm being disconnected from the grid due to the crowbar protection action is calculated; Under the low voltage ride-through operation strategy, the terminal voltage of each DFIG is calculated, and the next DFIG to be disconnected and the time of disconnection are determined; after the topology structure changes due to the unit being disconnected, the impedance matrix is ​​corrected, and the next DFIG to be disconnected and the time of disconnection are re-determined until all DFIGs are disconnected; Considering the switch from the conventional operation strategy to the low voltage ride-through operation strategy, the probability distribution V of two consecutive off-grid events is established based on the transfer matrix T. m and V m+1 Based on the probability of each DFIG in the wind farm being disconnected from the grid due to the crowbar protection action, the initial state probability distribution V1 and the transfer matrix T are solved, and then the state probability distribution of each disconnection event is obtained; Assuming there are n DFIGs, the optional state set E when each event occurs is defined as: E={S1,…,S k ,…,S i ,…,S n ,S n+1 }; where S1,…,Sn represent the disconnection of the first DFIG to the nth DFIG, S n+1 Indicates that all DFIGs are off-grid; the state is divided into absorption state and transfer state; when in the absorption state, the transfer probability to other states is 0, and it will stay in this state; when in the transfer state, it will transfer to other states according to the transfer probability when the next event occurs; state S1,…,S n is the transition state, state S n+1 It is in the absorption state; Based on the probability of each DFIG in the wind farm being disconnected from the grid due to the crowbar protection action, the initial state probability distribution V1 is solved, which is: The direction close to the PCC point is described as the front side, and the direction away from the PCC point is described as the back side; When the initial event is in the transition state S1,…,S n When the state S k The probability P k (1) After the strategy is switched, the DFIG that is closer to the PCC point will be disconnected from the grid earlier. k (1) This is equivalent to the probability that all DFIGs in front of the kth DFIG trigger the crowbar protection, while the kth DFIG does not trigger the crowbar protection, that is, q k is the probability of the kth DFIG being disconnected from the grid due to the crowbar protection action; When the initial event is in the absorbing state S n+1 When In summary, the state probability distribution V1 of the initial event is described as:

2. A chain path analysis method in a doubly-fed wind farm considering strategy switching as claimed in claim 1, characterized in that: Under the conventional operation strategy, the probability of each DFIG in the wind farm being disconnected from the grid due to the crowbar protection action is calculated as follows: Among them, P m1 , P m2 They are the minimum power and maximum power of DFIG when operating at MPPT; P c is the critical active power of DFIG for crowbar protection action, specifically: f P (P) is the probability density function of the active power output when DFIG operates in MPPT, specifically: in, ρ is the air density; S w is the windward swept area of ​​the wind turbine blades; C pmax is the maximum wind energy utilization coefficient; opt is the optimal tip speed ratio; n p is the number of DFIG pole pairs; N is the gearbox speed ratio; P sc is the critical stator active power when the crowbar protection is actuated; R w is the radius of the wind wheel; ω1 is the synchronous angular frequency; k is the shape parameter; c is the scale parameter; P is the DFIG active power.

3. A chain path analysis method in a doubly-fed wind farm considering strategy switching as claimed in claim 1, characterized in that: Under the low voltage ride-through operation strategy, the terminal voltage of each DFIG is calculated as follows: U=U p +ZI; Among them, U and I are matrices composed of the terminal voltage and output reactive current of each DFIG respectively; U p is the matrix composed of PCC point voltages; Z is the impedance matrix formed by the branch addition method according to the wind farm topology. When the topology changes due to the unit being disconnected from the grid, the impedance matrix Z needs to be recalculated.

4. A chain path analysis method in a doubly-fed wind farm considering strategy switching as claimed in claim 1, characterized in that: According to the relationship between the DFIG off-grid time t and the terminal voltage, the next off-grid DFIG and off-grid time are determined as follows: The relationship between the DFIG off-grid moment and the terminal voltage is described as: t = 1.96U t +0.23(0.2≤U t ≤0.9); Among them, U t It is the per unit voltage of wind turbine terminal.

5. The chain path analysis method in a doubly-fed wind farm considering strategy switching according to claim 1, characterized in that: Under the low voltage ride-through operation strategy, the chain disconnection of multiple DFIGs in the field is manifested as: sequential disconnection from the point close to the PCC point to the point far away from the PCC point, and the disconnection speed is getting faster and faster, showing an accelerated development.

6. A chain path analysis method in a doubly-fed wind farm considering strategy switching as claimed in claim 1, characterized in that: Based on the transfer matrix T, the probability distribution V of two consecutive off-line events is established m and V m+1 The deductive relationship is as follows: TV m =V m+1 ; Among them, V m 、V m+1 are the state probability distributions of the mth and m+1th events respectively.

7. A chain path analysis method in a doubly-fed wind farm considering strategy switching as claimed in claim 1, characterized in that: Based on the probability of each DFIG in the wind farm being disconnected from the grid due to the crowbar protection action, the transfer matrix T is solved, which is: Solve for the transition state S k , k≤n The probability of transferring to various states in the next event: Since the transfer state will not be transferred to the state of the next event again, and after the strategy switching, the DFIG is disconnected from the grid in sequence from the electrical distance to the PCC point from near to far, the state S k It is impossible to transfer to the state S1,…,S of the next event k , that is, P 1k ,…,P kk All are 0; By state S k Transfer to state S in the next event i Probability P ik It is equivalent to the probability that all other DFIGs between the kth DFIG and the ith DFIG trigger the crowbar protection and the ith DFIG does not trigger the crowbar protection, that is, By state S k Transfer to the next event absorbing state S n+1 The probability P n+1,k It is equivalent to the probability that all DFIGs behind the kth DFIG trigger the crowbar protection, that is, When the nth DFIG at the end of the collector line is disconnected from the grid, all DFIGs in the wind farm are also disconnected from the grid, and the state must be changed from S n Transfer to state S in the next event n+1 , that is, P n+1,n =1; When the event is in the absorbing state S n+1 When P n+1,n+1 =1; The transfer matrix T is described as:

8. A chain path analysis system in a doubly-fed wind farm considering strategy switching, based on a chain path analysis method in a doubly-fed wind farm considering strategy switching as claimed in any one of claims 1 to 7, characterized in that: include: The DFIG off-grid analysis module in the conventional operation strategy stage is used to calculate the probability of each DFIG in the wind farm being off-grid due to the crowbar protection action under the conventional operation strategy; The DFIG off-grid analysis module in the low voltage ride-through operation strategy stage is used to calculate the terminal voltage of each DFIG under the low voltage ride-through operation strategy, and determine the next off-grid DFIG and the off-grid time; after the topology structure changes due to the unit off-grid, the impedance matrix is ​​corrected, and the next off-grid DFIG and the off-grid time are re-determined until all DFIGs are off-grid; The off-grid analysis module in the DFIG operation strategy switching phase is used to consider the switch from the conventional operation strategy to the low voltage ride-through operation strategy. Based on the transfer matrix T, the probability distribution V of the state of two consecutive off-grid events is established. m and V m+1 Based on the probability of each DFIG in the wind farm being disconnected from the grid due to the crowbar protection action, the initial state probability distribution V1 and the transfer matrix T are solved, and then the state probability distribution of each disconnection event is obtained.

9. A terminal device, comprising a processor and a memory, wherein the processor is used to implement instructions; and the memory is used to store multiple instructions, characterized in that: The instructions are suitable for being loaded by a processor and executing the chain path analysis method within a doubly-fed wind farm considering strategy switching as described in any one of claims 1-7.

Images