A method and system for identifying excitation inrush current and fault current of a double-fed wind farm sending-out transformer

Abstract

The application discloses a kind of excitation inrush current and fault current identification method of double-fed wind farm sending-out transformer in the field of transformer relay protection, to solve the technical problem of differential protection delayed action or even refusal to operate.The method comprises: obtaining the current of each phase of the star side of the transformer and the current of each phase of the delta side;The current of each phase of the star side of the transformer and the current of each phase of the delta side are converted to the same reference side, and the differential current is calculated;The differential current is compared with the preset protection starting element setting value, if the differential current is not greater than the preset protection starting element setting value, it is determined that the normal operation or the protection starting condition is not reached;If the differential current is greater than the preset protection starting element setting value, the excitation inrush current and fault current identification process is started.The application can accurately distinguish the excitation inrush current and the fault current, resist frequency offset interference, and improve the speed and reliability of the transformer main protection under the new energy scene.

Description

Technical Field

[0001] This invention relates to a method and system for identifying inrush current and fault current of a doubly fed wind farm output transformer, belonging to the field of transformer relay protection technology. Background Technology

[0002] In recent years, the steady development of power electronic equipment control technology has laid a solid foundation for the widespread application of variable-speed constant-frequency wind turbines, significantly promoting the development of wind power technology. Among them, doubly-fed induction generator (DFIG) wind turbines have become the mainstream wind turbine type due to their high operational reliability and excellent variable-speed constant-frequency power generation characteristics. Furthermore, with the rapid development of the wind power industry, most grid-connected wind farms currently possess low-voltage ride-through capabilities, enabling them to maintain grid connection during grid faults. To achieve the national "dual-carbon" strategic goal, wind power installed capacity continues to grow. Due to the significant regional concentration of wind energy resources in my country, most wind farms adopt a large-scale centralized grid connection method, connecting to the grid through the main transformer of the collection station. Currently, the relay protection of the main transformer at the collection station mostly adopts traditional power transformer protection schemes.

[0003] However, doubly-fed induction generators (DFIGs) differ significantly from traditional synchronous generators in control technology and grid connection methods, resulting in different transient electrical characteristics of DFIGs during grid faults. When a severe fault occurs, the terminal voltage of the DFIG drops sharply, and the rotor current rises rapidly. Once the crowbar protection threshold is exceeded, the crowbar circuit engages, the rotor-side converter is locked, and the rotor circuit voltage drops to zero. At this time, the harmonic components of the fault short-circuit current increase significantly and experience frequency shift, causing inaccurate calculation of the second harmonic proportion in the main transformer differential protection. Traditional second harmonic criteria cannot accurately identify the inrush current, leading to delayed or even non-operational differential protection. Due to the support of the grid voltage, the voltage frequency can be maintained at the power frequency level. This creates a discrepancy between voltage and current frequencies, interfering with the analytical and judgment capabilities of relay protection devices. Therefore, traditional transformer differential protection schemes are not well adapted to the fault characteristics of large-scale wind farms and cannot meet the requirements of new power systems for the selectivity, speed, sensitivity and reliability of relay protection. Therefore, it is urgent to study and develop new protection methods suitable for wind farm transmission transformers. Summary of the Invention

[0004] The purpose of this invention is to provide a method and system for identifying inrush current and fault current of the sending transformer in a doubly-fed wind farm. This method can overcome the limitations of traditional second harmonic inrush current criteria in doubly-fed wind turbine grid-connected systems, solve the problem of delayed operation or even failure to operate of differential protection, and ensure safe and stable operation when a high proportion of new energy sources are connected to the grid.

[0005] To solve the above-mentioned technical problems, the present invention is implemented using the following technical solution.

[0006] In a first aspect, the present invention provides a method for identifying inrush current and fault current of a doubly-fed wind farm's output transformer, comprising:

[0007] Obtain the phase currents on the star-connected side and the phase currents on the delta-connected side of the transformer;

[0008] The differential current is calculated by converting the phase currents on the star side and the phase currents on the delta side of the transformer to the same reference side.

[0009] The differential current is compared with the preset protection starting element setting value. If the differential current is not greater than the preset protection starting element setting value, it is determined to be normal operation or protection starting conditions not met. If the differential current is greater than the preset protection starting element setting value, the excitation inrush current and fault current identification process is initiated.

[0010] In conjunction with the first aspect, the calculation expression for the differential current further includes:

[0011] ;

[0012] in, Indicates differential current; This represents the current after conversion on the star-connected side of the transformer; This represents the current after deflection on the delta side of the transformer.

[0013] In conjunction with the first aspect, the initiation excitation inrush current and fault current identification process further includes:

[0014] Obtain the phase voltages on the star side of the transformer ;

[0015] The equivalent circuit of a three-phase transformer is constructed based on the star-delta connection, and the three-phase circuit equation is derived using the phase voltages on the star side and the phase currents on the delta side of the transformer.

[0016] The three-phase circuit equations are simplified and discretized to obtain the three-phase magnetizing inductance of the transformer.

[0017] The three-phase magnetizing inductance of the transformer is estimated online using the adaptive forgetting factor least squares method to obtain the online estimated value of the magnetizing inductance at each sampling time.

[0018] Construct an excitation inductance waveform curve with time as the x-axis and the online estimated value of the excitation inductance as the y-axis. ;

[0019] Calculate the waveform curve of the excitation inductor The area enclosed by the preset baseline and peak inductance within a fixed period The area of ​​the rectangle enclosed by the horizontal line and the preset baseline. ;

[0020] Calculate the area and the area of ​​a rectangle The ratio of the two values ​​yields the area coefficient. and the area coefficient With preset protection activation threshold Comparison, if the area coefficient Less than the preset protection activation threshold If the current flowing through the transformer is determined to be an inrush current, the transformer protection is locked out; if the area coefficient is... Not less than the preset protection start threshold If so, the current flowing through the transformer is determined to be the fault current, and the asynchronous method is used to further distinguish the fault type.

[0021] In conjunction with the first aspect, the expression for the three-phase magnetizing inductance of the transformer is further as follows:

[0022] ;

[0023] in, This represents the magnetizing inductance of phase A of the transformer at the nth sampling point; This represents the B-phase magnetizing inductance of the transformer at the nth sampling point; This represents the C-phase magnetizing inductance of the transformer at the nth sampling point; This represents the voltage of phase A on the star-connected side of the transformer at the nth sampling point; This represents the A-phase voltage on the star-connected side of the transformer at the (n+1)th sampling point; This represents the B-phase voltage on the star-connected side of the transformer at the nth sampling point; This represents the B-phase voltage on the star-connected side of the transformer at the (n+1)th sampling point; This represents the C-phase voltage on the star-connected side of the transformer at the nth sampling point; This represents the C-phase voltage on the star-connected side of the transformer at the (n+1)th sampling point; Represents the nth sampling point Discrete values; Represents the (n+1)th sampling point Discrete values; Represents the nth sampling point Discrete values; Represents the (n+1)th sampling point Discrete values; Represents the nth sampling point Discrete values; Represents the (n+1)th sampling point Discrete values.

[0024] In conjunction with the first aspect, further, the adaptive forgetting factor least squares method is used to estimate the three-phase magnetizing inductance of the transformer online, obtaining the online estimated values ​​of the magnetizing inductance at each sampling time, including:

[0025] Step S51: Linearly convert the three-phase magnetizing inductance of the transformer to obtain the output at the current moment;

[0026] Step S52: Before starting the recursion, the relevant parameters of the adaptive forgetting factor least squares method are initialized; wherein, the initialization of the relevant parameters includes setting the initial parameter estimates, the initial covariance matrix, and the initial value of the forgetting factor.

[0027] Step S53: Perform recursive calculations sequentially at each sampling time, using the parameter estimate from the previous time and the output at the current time to calculate the prediction error at the current time;

[0028] Step S54: Dynamically adjust the forgetting factor based on the prediction error at the current moment;

[0029] Step S55: Based on the adjusted forgetting factor, calculate the gain matrix, and correct the online estimate of the excitation inductance at the current moment using the gain matrix and the current prediction error to obtain the corrected online estimate of the excitation inductance.

[0030] Step S56: Update the covariance matrix based on the adjusted forgetting factor and gain matrix;

[0031] Step S57: Pass the covariance matrix at the current time to the recursive calculation at the next time and repeat steps S53 to S56 to obtain the corrected online estimate of the excitation inductance at each sampling time.

[0032] In conjunction with the first aspect, the prediction error at the current moment is further expressed as:

[0033] ;

[0034] in, This represents the prediction error for the nth sampling point; This represents the output value of the nth sampling point; This represents the regression value at the nth sampling point; This represents the parameter estimate for the (n-1)th sampling point.

[0035] In conjunction with the first aspect, the expression for correcting the current online estimate of the excitation inductance is further as follows:

[0036] ;

[0037] in, This represents the corrected online estimate of the magnetizing inductance; This represents the online estimate of the excitation inductance before correction; Represents the gain matrix; This represents the prediction error for the nth sampling point.

[0038] In conjunction with the first aspect, the expression for the area coefficient is further as follows:

[0039] ;

[0040] ;

[0041] ;

[0042] in, This represents the area enclosed by the excitation inductance waveform curve and the preset baseline. This represents the area of ​​the rectangle enclosed by the horizontal line where the inductance peak value is located within a fixed period and the preset baseline. This represents the number of sampling points within one period; Indicates the index number of the sampling point; This represents the online estimate of the magnetizing inductance at the i-th sampling point; Indicates the baseline; Indicates the time step; Indicates the peak value of the magnetizing inductance; Indicates the length of a period; This represents the area coefficient.

[0043] In conjunction with the first aspect, the asynchronous method is further used to further distinguish fault types, including:

[0044] The fault type is distinguished by comparing the timing relationship between the differential current surge and the voltage drop. If the differential current surge lags behind the voltage drop, it is judged as an external fault.

[0045] If the sudden increase in differential current occurs simultaneously with the sudden drop in voltage, then it is determined that a fault has occurred inside the transformer.

[0046] Secondly, a system for identifying inrush current and fault current of a doubly-fed wind farm's output transformer includes:

[0047] A data acquisition model is used to acquire the phase currents on the star side and the phase currents on the delta side of the transformer.

[0048] The current calculation module is used to convert the phase currents of the star side and the phase currents of the delta side to the same reference side and calculate the differential current.

[0049] The fault identification module is used to compare the differential current with the preset protection starting element setting value. If the differential current is not greater than the preset protection starting element setting value, it is determined to be normal operation or not reached; if the differential current is greater than the preset protection starting element setting value, the excitation inrush current and fault current identification process is initiated.

[0050] Compared with the prior art, the beneficial effects achieved by the present invention are as follows:

[0051] This invention acquires the phase currents and voltages on the star-connected side of the transformer and the phase currents on the delta-connected side. It calculates the differential current by converting the currents on both sides to the same reference side. The differential current is compared with a preset setting value: if it does not exceed the setting value, it is determined to be normal operation or an external fault; if it exceeds the setting value, the identification process is initiated: first, based on the constructed transformer equivalent circuit model, the three-phase excitation inductance value is solved in real time using the adaptive forgetting factor least squares method to overcome the interference of frequency offset and non-periodic components; then, within a fixed time window, the area ratio between the inductance waveform and the set baseline is quantified to generate an area coefficient characterizing the inductance fluctuation characteristics; if the area coefficient is lower than the protection threshold, it is determined to be an inrush current and the protection is blocked; if it reaches or exceeds the threshold, it is determined to be a fault condition. Furthermore, the time synchronization of differential current surges and voltage drops is combined to accurately distinguish between internal faults (immediate action when synchronously occurring) and external CT saturation (blocking when differential current lags).

[0052] This method significantly improves the accuracy and speed of the main protection operation of the doubly fed wind farm's transmitting transformer, effectively solves the problem of delay or failure to operate caused by the failure of traditional harmonic criteria, and ensures the stable operation of the power grid under the high proportion of new energy access. Attached Figure Description

[0053] Figure 1 The diagram shown is a flowchart of a method for identifying inrush current and fault current of a doubly fed wind farm transmission transformer according to an embodiment of the present invention.

[0054] Figure 2 The figure shown is a PSCAD simulation model diagram of a doubly fed wind turbine grid-connected system provided in an embodiment of the present invention;

[0055] Figure 3 The diagram shown is a wiring diagram of a doubly fed wind turbine grid-connected system according to an embodiment of the present invention;

[0056] Figure 4 The diagram shown is an equivalent circuit diagram of a Y-Δ connected three-phase transformer according to an embodiment of the present invention.

[0057] Figure 5 The figure shown is a current and voltage waveform analysis diagram of a three-phase short-circuit fault on the wind turbine side provided in an embodiment of the present invention;

[0058] Figure 6 The figure shown is an analysis diagram of the inductance amplitude waveform during a three-phase short-circuit fault on the wind turbine side, provided by an embodiment of the present invention.

[0059] Figure 7 The figure shown is an area factor analysis diagram for three-phase short-circuit faults on the wind turbine side provided by an embodiment of the present invention;

[0060] Figure 8 The figure shown is a current and voltage waveform analysis diagram of a phase-to-phase short-circuit fault on the wind turbine side provided by an embodiment of the present invention;

[0061] Figure 9 The figure shown is an analysis diagram of the inductance amplitude waveform during a phase-to-phase short-circuit fault on the wind turbine side, provided by an embodiment of the present invention.

[0062] Figure 10 The figure shown is an area factor analysis diagram for phase-to-phase short-circuit faults on the wind turbine side provided by an embodiment of the present invention;

[0063] Figure 11 The figure shown is a current and voltage waveform analysis diagram of an inter-turn fault on the system side provided by an embodiment of the present invention;

[0064] Figure 12 The figure shown is an analysis diagram of the inductor amplitude waveform during an inter-turn fault on the system side provided in an embodiment of the present invention;

[0065] Figure 13 The figure shown is an area factor analysis diagram for inter-turn faults on the system side provided by an embodiment of the present invention;

[0066] Figure 14 The figure shown is an analysis diagram of the excitation inrush current and voltage waveforms provided in an embodiment of the present invention;

[0067] Figure 15 The figure shown is a waveform analysis diagram of the inrush current amplitude of the inductor provided in an embodiment of the present invention;

[0068] Figure 16 The figure shown is an area factor analysis diagram for inrush current provided by an embodiment of the present invention;

[0069] Figure 17 The figure shown is an analysis diagram of the inter-turn fault current and voltage waveforms provided in an embodiment of the present invention.

[0070] Figure 18 The figure shown is a waveform analysis diagram of the amplitude of the inter-turn fault inductor airdropped to the system side according to an embodiment of the present invention;

[0071] Figure 19The figure shown is an area factor analysis diagram for inter-turn faults airdropped into the system side, provided by an embodiment of the present invention. Detailed Implementation

[0072] The technical solution of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present invention and the specific features in the embodiments are detailed descriptions of the technical solution of the present invention, rather than limitations thereof. In the absence of conflict, the embodiments of the present invention and the technical features in the embodiments can be combined with each other.

[0073] The term "and / or" simply describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. Additionally, the character " / " generally indicates that the preceding and following related objects have an "or" relationship. Example 1

[0074] join Figure 1 This embodiment introduces a method for identifying inrush current and fault current of a doubly-fed wind farm's output transformer, characterized by the following steps:

[0075] Step S1: Obtain the current of each phase on the star-connected side of the transformer. and the phase currents on the delta side of the transformer ;

[0076] Step S2: Calculate the differential current by converting the phase currents on the star-connected side and the phase currents on the delta-connected side of the transformer to the same reference side (e.g., the star-connected side). ;

[0077] The calculation expression for the differential current includes:

[0078] ;

[0079] in, Indicates differential current; This represents the current after conversion on the star-connected side of the transformer; This represents the current after deflection on the delta side of the transformer.

[0080] Step S3: Compare the differential current with the preset protection start-up element setting value. If the differential current is less than or equal to the preset protection starting element setting value, the comparison is made. If the differential current is greater than the preset protection start-up setting value, it is determined to be either in normal operation or the protection start-up condition has not been met. If so, the inrush current and fault current identification process will be initiated.

[0081] It should be noted that the protection start threshold The value range is (0.15~0.2). In the formula This represents the rated current on the transformer reference side. In this embodiment of the invention, the protection initiation threshold is... The value is 0.2 .

[0082] For details, see Figure 4 The inrush current and fault current identification process includes the following steps:

[0083] Step S31: Obtain the voltage of each phase on the star-connected side of the transformer. ;

[0084] Step S32: Construct the equivalent circuit of the three-phase transformer based on the star-side and delta-side connection method (Y-Δ connection method), and derive the three-phase loop equation using the phase voltage of each phase on the star side and the phase current of each phase on the delta side of the transformer.

[0085] The expression for the three-phase circuit equation is as follows:

[0086] (2)

[0087] in, This indicates the voltage of phase A on the star-connected side of the transformer; This indicates the voltage of phase B on the star-connected side of the transformer; This indicates the C-phase voltage on the star-connected side of the transformer; This indicates the resistance of phase A of the transformer; This indicates the resistance of phase B of the transformer; This indicates the resistance of phase C of the transformer; This indicates the A-phase current on the star-connected side of the transformer; This indicates the B-phase current on the star-connected side of the transformer; This indicates the C-phase current on the star-connected side of the transformer; This indicates the leakage inductance of phase A of the transformer; This indicates the leakage inductance of phase B of the transformer; This indicates the leakage inductance of phase C of the transformer; This indicates the excitation resistance of phase A of the transformer; This indicates the magnetizing resistance of phase B of the transformer; This indicates the excitation resistance of phase C of the transformer; This represents the current in phase A winding on the delta side of the transformer; This represents the current in the B-phase winding on the delta side of the transformer; This represents the current in the C-phase winding on the delta side of the transformer; This indicates the circulating current in the delta winding; This represents the magnetizing inductance of phase A of the transformer; This represents the magnetizing inductance of phase B of the transformer; This represents the magnetizing inductance of phase C of the transformer; Indicates time.

[0088] Step S33: Simplify and discretize the three-phase circuit equation obtained from formula (2) to obtain the three-phase magnetizing inductance of the transformer;

[0089] The expression for the three-phase magnetizing inductance of the transformer is:

[0090] (3)

[0091] in, This represents the magnetizing inductance of phase A of the transformer at the nth sampling point; This represents the B-phase magnetizing inductance of the transformer at the nth sampling point; This represents the C-phase magnetizing inductance of the transformer at the nth sampling point; This represents the voltage of phase A on the star-connected side of the transformer at the nth sampling point; This represents the A-phase voltage on the star-connected side of the transformer at the (n+1)th sampling point; This represents the B-phase voltage on the star-connected side of the transformer at the nth sampling point; This represents the B-phase voltage on the star-connected side of the transformer at the (n+1)th sampling point; This represents the C-phase voltage on the star-connected side of the transformer at the nth sampling point; This represents the C-phase voltage on the star-connected side of the transformer at the (n+1)th sampling point; Represents the nth sampling point Discrete values; Represents the (n+1)th sampling point Discrete values; Represents the nth sampling point Discrete values; Represents the (n+1)th sampling point Discrete values; Represents the nth sampling point Discrete values; Represents the (n+1)th sampling point Discrete values.

[0092] Furthermore, the discrete values ​​can be rewritten as the following expression:

[0093] (4)

[0094] in, Represents the (n-1)th sampling point Discrete values; Represents the (n-1)th sampling point Discrete values; Represents the (n-1)th sampling point Discrete values; Represents the (n+2)th sampling point Discrete values; Represents the (n+2)th sampling point Discrete values; Represents the (n+2)th sampling point Discrete values; This indicates the duration of the power frequency cycle.

[0095] Step S34: Use the adaptive forgetting factor least squares method to estimate the three-phase magnetizing inductance of the transformer to obtain the online estimated value of the magnetizing inductance at each time.

[0096] Specifically, the process of obtaining the online estimate of the excitation inductance at each moment includes:

[0097] Step S341: Linearly convert the three-phase magnetizing inductance of the transformer to obtain the output at the current moment;

[0098] Specifically, the derived expression for the three-phase magnetizing inductance of the transformer, rewritten in linear parametric form, is as follows:

[0099] (5)

[0100] in, Indicates the output quantity; Indicates the regression quantity; This represents the parameter to be estimated.

[0101] Taking phase A as an example, the output, regression, and parameters to be estimated are expressed as follows:

[0102] (6)

[0103] (7)

[0104] (8)

[0105] Step S342: Before starting the recursion, initialize the relevant parameters of the adaptive forgetting factor least squares method; wherein, initializing the relevant parameters includes setting initial parameter estimates. Initial covariance matrix and the initial value of the forgetting factor ;

[0106] Step S343: Perform recursive calculations sequentially at each sampling time, using the parameter estimate from the previous time and the output at the current time to calculate the prediction error at the current time;

[0107] The expression for the current prediction error is:

[0108] (9)

[0109] in, This represents the prediction error for the nth sampling point; This represents the output value of the nth sampling point; This represents the regression value at the nth sampling point; This represents the parameter estimate for the (n-1)th sampling point.

[0110] Step S344: Based on the prediction error at the current moment, adaptively and dynamically adjust the forgetting factor to obtain the adjusted forgetting factor;

[0111] The adjusted forgetting factor expression is as follows:

[0112] (10)

[0113] in, This represents the minimum value of the forgetting factor, and ; This represents the decay parameter of the forgetting factor; Represents the natural constant.

[0114] Here, the error is mapped to the forgetting factor range by an exponential function, achieving an adaptive balance between enhancing parameter tracking capability when the error is large and improving steady-state accuracy when the error is small.

[0115] Step S345: Based on the adjusted forgetting factor, calculate the gain matrix, the expression of which is:

[0116] (11)

[0117] in, Represents the gain matrix; This represents the covariance matrix of the (n-1)th sampling point; This represents the regression value at the nth sampling point; This represents the forgetting factor at the nth sampling point.

[0118] Furthermore, the online estimate of the excitation inductance at the current moment is corrected by the gain matrix and the prediction error at the current moment, and its expression is:

[0119] (12)

[0120] in, This represents the corrected online estimate of the magnetizing inductance; This represents the online estimate of the excitation inductance before correction; This represents the gain matrix of the nth sampling point; This represents the prediction error for the nth sampling point.

[0121] Step S346: Update the covariance matrix based on the adjusted forgetting factor and gain matrix;

[0122] As the online estimate of the excitation inductance is continuously updated, the covariance matrix gradually changes. The expression for this covariance matrix is:

[0123] (13)

[0124] in, Let denote the covariance matrix; E denotes the identity matrix.

[0125] Step S347: Pass the covariance matrix at the current time to the recursive calculation at the next time and repeat steps S343 to S346 to obtain the corrected online estimate of the excitation inductance at each sampling time.

[0126] Step S35: Construct an excitation inductance waveform curve with time as the x-axis and the online estimated value of the excitation inductance as the y-axis. Calculate the excitation inductance waveform curve The area enclosed by the preset baseline y=-a Peak inductance within a fixed period The area of ​​the rectangle enclosed by the horizontal line and the preset baseline. The value of baseline 'a' ranges from 2 to 3 times the excitation inductance of the transformer during normal operation. In this embodiment of the invention, it is taken as 2 times, and its area coefficient is expressed as follows:

[0127] ;

[0128] ;

[0129] ;

[0130] in, This represents the area enclosed by the excitation inductance waveform curve and the preset baseline. This represents the area of ​​the rectangle enclosed by the horizontal line where the inductance peak value is located within a fixed period and the preset baseline. This represents the number of sampling points within one period; Indicates the index number of the sampling point; This represents the online estimate of the magnetizing inductance at the i-th sampling point; Indicates the baseline; Indicates the time step; Indicates the peak value of the magnetizing inductance; Indicates the length of a period; This represents the area coefficient.

[0131] Step S36, Calculation and The ratio of the two values ​​yields the area coefficient. and area coefficient With preset protection activation threshold Comparison, if area coefficient Less than the preset protection activation threshold If the current flowing through the transformer is determined to be an inrush current, the transformer protection will be locked to prevent malfunction of the protection device; if the area coefficient is... Greater than or equal to the preset protection activation threshold If the fault current is determined to be the fault current, the asynchronous method is used to further distinguish the fault type, that is, to distinguish between faults within the zone and faults outside the zone, so as to take precise protection measures.

[0132] Among them, the protection start threshold The value ranges from 0.3 to 0.5, and in this embodiment of the invention, it is the protection activation threshold. The value is 0.3.

[0133] Furthermore, the asynchronous method is used to further distinguish fault types, including:

[0134] The timing relationship between the differential current surge and the voltage drop is compared to distinguish the fault type. If the differential current surge lags behind the voltage drop, it is determined that the current transformer (CT) is saturated due to an external fault, and the transformer protection device will not operate in this case.

[0135] If the sudden increase in differential current occurs simultaneously with the sudden drop in voltage, it is determined that an internal fault has occurred in the transformer (internal fault). At this time, the transformer protection device will act in a timely manner to protect the transformer from damage.

[0136] Finally, simulations revealed the following characteristics in the variation of transformer magnetizing inductance and area factor under different operating conditions:

[0137] During inrush current, the value of the magnetizing inductance fluctuates between a large and a small value, at which point the area coefficient... Smaller;

[0138] When an intra-zone fault occurs, the excitation inductance value is small (approaching 0) and does not fluctuate, and the area factor... It is relatively large, and approaches 1;

[0139] During normal operation and external faults, the excitation inductance value is large and does not fluctuate, and the area coefficient is... It is relatively large, and approaches 1;

[0140] When a fault outside the zone causes CT saturation, the excitation inductance value will also fluctuate, affecting the area coefficient. The differential current is relatively small, but under this operating condition, the timing of the differential current change is significantly delayed compared to the timing of the voltage change. Under other operating conditions, the differential current occurs synchronously. Based on this, the inrush current differential current can be distinguished from the fault differential current. Example 2

[0141] This embodiment utilizes PSCAD to simulate a doubly-fed induction generator (DFIG) grid-connected wind turbine system, and builds a corresponding simulation model, such as... Figure 2 As shown in the diagram. Taking a 150MVA doubly-fed wind farm as an example, this wind farm consists of 30 5MVA doubly-fed wind turbines connected in parallel. The doubly-fed wind turbine unit wiring is as follows: each doubly-fed wind turbine has a terminal voltage of 0.69kV, which is connected to a 35kV busbar via a prefabricated transformer. The voltage is then stepped up to 110kV before being connected to the grid. The system capacity is 500MVA. The prefabricated transformer has a capacity of 6MVA, with a Yd wiring configuration and a leakage reactance of 6%. The main transformer has a capacity of 160MVA, with a YNd wiring configuration and a leakage reactance of 10%. For specific wiring details, please refer to [link to details]. Figure 3 The parameters of the doubly fed wind turbine are shown in Table 1.

[0142] Table 1

[0143]

[0144] Simulations were performed on the main transformer under the following operating conditions: three-phase short-circuit fault on the wind turbine side, phase-to-phase short-circuit fault on the wind turbine side, inter-turn fault and inrush current on the system side, and inter-turn fault caused by no-load operation. In this embodiment, the internal fault occurred at 1.5s and lasted for 0.2s; no-load operation occurred at 0.2s. Simulations of inrush current and inter-turn fault caused by no-load operation were performed. The specific implementation steps were the same as in Embodiment 1. The simulation results of the three-phase short-circuit fault on the wind turbine side are as follows: Figure 5 ~ Figure 7 Simulation results of phase-to-phase short-circuit fault on the wind turbine side are as follows: Figure 8 ~ Figure 10 Simulation results of inter-turn faults on the system side are as follows: Figure 11 ~ Figure 13 The simulation waveform of inrush current is as follows: Figure 14 ~ Figure 16 Simulation results for the system-side inter-turn fault condition airdropped as follows: Figure 17 ~ Figure 19 .

[0145] Simulation examples show that when a three-phase short-circuit fault or a phase-to-phase short-circuit fault occurs on the wind turbine side, the excitation inductance value drops significantly and approaches 0. At this time, the calculated area coefficient approaches 1 and is greater than the protection start threshold, and does not meet the conditions of the asynchronous method. At this time, the transformer protection device will operate.

[0146] When a system-side inter-turn fault occurs, the excitation inductance value drops significantly and becomes relatively stable. At this time, the calculated area factor approaches 1, which is greater than the protection start threshold and does not meet the conditions of the asynchronous method, so the transformer protection device operates.

[0147] When the transformer is closed under no-load (no-load), an inrush current is generated. At this time, the magnetizing inductance fluctuates between a large and a small value with a large amplitude. The calculated area coefficient is less than the protection start threshold, and the transformer protection device is locked.

[0148] When an inter-turn fault occurs on the system side, the excitation inductance value is small and relatively stable. At this time, the calculated area coefficient is close to 1, which is greater than the protection start threshold and does not meet the asynchronous method criterion, so the transformer protection operates. Example 3

[0149] A system for identifying inrush current and fault current of a doubly-fed wind farm's output transformer includes:

[0150] A data acquisition model is used to acquire the phase currents on the star side and the phase currents on the delta side of the transformer.

[0151] The current calculation module is used to convert the phase currents of the star side and the phase currents of the delta side to the same reference side and calculate the differential current.

[0152] The fault identification module is used to compare the differential current with the preset protection starting element setting value. If the differential current is not greater than the preset protection starting element setting value, it is determined to be normal operation or external fault. If the differential current is greater than the preset protection starting element setting value, the excitation inrush current and fault current identification process is initiated.

[0153] This system directly calculates the excitation inductance using the recursive least squares method, completely avoiding the shortcomings of traditional second harmonic criterion which is affected by frequency offset (such as harmonic distortion caused by the operation of the doubly-fed wind turbine's crowbar). An adaptive forgetting factor mechanism dynamically optimizes the weights of historical data, enabling the algorithm to quickly track inductance fluctuations during inrush current transients and maintain high accuracy during steady-state processes. An area coefficient criterion quantifies the characteristics of inductance fluctuations, solving the problem of distinguishing between fault current and inrush current in renewable energy scenarios. The timing verification stage effectively eliminates the risk of false tripping caused by CT saturation due to external faults through millisecond-level differential current-voltage surge timing analysis. Ultimately, this achieves high accuracy in fault identification within the fault zone and reliable excitation inrush current blocking, avoiding false tripping of traditional protection systems in wind turbine grid-connected scenarios.

[0154] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of the present invention without departing from the spirit and scope of the claims. All of these forms are within the protection scope of the present invention.

Claims

1. A method for identifying inrush current and fault current in a doubly-fed wind farm's transmitting transformer, characterized in that, include: Obtain the phase currents on the star side and the phase currents on the delta side of the transformer; The differential current is calculated by converting the currents of each phase on the star side and the currents of each phase on the delta side of the transformer to the same reference side. The expression for the differential current is as follows: ； in, Indicates differential current; This represents the current after conversion on the star-connected side of the transformer; This represents the current after delta conversion on the transformer side; The differential current is compared with a preset protection starting element setting value. If the differential current is not greater than the preset protection starting element setting value, it is determined to be normal operation or the protection starting condition has not been met. If the differential current is greater than the preset protection starting element setting value, the inrush current and fault current identification process is initiated. The inrush current and fault current identification process includes: Obtain the phase voltages on the star side of the transformer ; The equivalent circuit of a three-phase transformer is constructed based on the star-delta connection, and the three-phase circuit equation is derived using the phase voltages on the star side and the phase currents on the delta side of the transformer. The three-phase circuit equations are simplified and discretized to obtain the three-phase magnetizing inductance of the transformer. The three-phase magnetizing inductance of the transformer is estimated online using the adaptive forgetting factor least squares method to obtain the online estimated value of the magnetizing inductance at each sampling time. Construct an excitation inductance waveform curve with time as the x-axis and the online estimated value of the excitation inductance as the y-axis. ; Calculate the waveform curve of the excitation inductor The area enclosed by the preset baseline and peak inductance within a fixed period The area of ​​the rectangle enclosed by the horizontal line and the preset baseline. ; Calculate the area and the area of ​​a rectangle The ratio of the two values ​​yields the area coefficient. and the area coefficient With preset protection activation threshold Comparison, if the area coefficient Less than the preset protection activation threshold If the current flowing through the transformer is determined to be an inrush current, the transformer protection is locked out; if the area coefficient is... Not less than the preset protection start threshold If the current flowing through the transformer is determined to be a fault current, an asynchronous method is used to further distinguish the fault type; wherein, the asynchronous method for further distinguishing the fault type includes: The fault type is distinguished by comparing the timing relationship between the differential current surge and the voltage drop. If the differential current surge lags behind the voltage drop, it is judged as an external fault. If the sudden increase in differential current occurs simultaneously with the sudden drop in voltage, then it is determined that a fault has occurred inside the transformer.

2. The method for identifying inrush current and fault current of a doubly-fed wind farm transmitting transformer according to claim 1, characterized in that, The expression for the three-phase magnetizing inductance of the transformer is: ； in, This represents the magnetizing inductance of phase A of the transformer at the nth sampling point; This represents the B-phase magnetizing inductance of the transformer at the nth sampling point; This represents the C-phase magnetizing inductance of the transformer at the nth sampling point; This represents the voltage of phase A on the star-connected side of the transformer at the nth sampling point; This represents the A-phase voltage on the star-connected side of the transformer at the (n+1)th sampling point; This represents the B-phase voltage on the star-connected side of the transformer at the nth sampling point; This represents the B-phase voltage on the star-connected side of the transformer at the (n+1)th sampling point; This represents the C-phase voltage on the star-connected side of the transformer at the nth sampling point; This represents the C-phase voltage on the star-connected side of the transformer at the (n+1)th sampling point; Represents the nth sampling point Discrete values; Represents the (n+1)th sampling point Discrete values; Represents the nth sampling point Discrete values; Represents the (n+1)th sampling point Discrete values; Represents the nth sampling point Discrete values; Represents the (n+1)th sampling point Discrete values.

3. The method for identifying inrush current and fault current of the output transformer in a doubly-fed wind farm according to claim 1, characterized in that, The three-phase magnetizing inductance of the transformer is estimated online using the adaptive forgetting factor least squares method, yielding online estimates of the magnetizing inductance at each sampling time, including: Step S51: Linearly convert the three-phase magnetizing inductance of the transformer to obtain the output at the current moment; Step S52: Before starting the recursion, the relevant parameters of the adaptive forgetting factor least squares method are initialized; wherein, the initialization of the relevant parameters includes setting the initial parameter estimates, the initial covariance matrix, and the initial value of the forgetting factor. Step S53: Perform recursive calculations sequentially at each sampling time, using the parameter estimate from the previous time and the output at the current time to calculate the prediction error at the current time; Step S54: Dynamically adjust the forgetting factor based on the prediction error at the current moment; Step S55: Based on the adjusted forgetting factor, calculate the gain matrix, and correct the online estimate of the excitation inductance at the current time using the gain matrix and the prediction error at the current time. Step S56: Update the covariance matrix based on the adjusted forgetting factor and gain matrix; Step S57: Pass the covariance matrix at the current time to the recursive calculation at the next time and repeat steps S53 to S56 to obtain the corrected online estimate of the excitation inductance at each sampling time.

4. The method for identifying inrush current and fault current of the doubly-fed wind farm transmitting transformer according to claim 3, characterized in that, The expression for the prediction error at the current moment is: ； in, This represents the prediction error for the nth sampling point; This represents the output value of the nth sampling point; This represents the regression value at the nth sampling point; This represents the parameter estimate for the (n-1)th sampling point.

5. The method for identifying inrush current and fault current of the output transformer in a doubly-fed wind farm according to claim 3, characterized in that, The expression for correcting the current online estimate of the excitation inductance is: ； in, This represents the corrected online estimate of the magnetizing inductance; This represents the online estimate of the excitation inductance before correction; Represents the gain matrix; This represents the prediction error for the nth sampling point.

6. The method for identifying inrush current and fault current of the output transformer in a doubly-fed wind farm according to claim 1, characterized in that, The expression for the area coefficient is: ； ； ； in, This represents the area enclosed by the excitation inductance waveform curve and the preset baseline. This represents the area of ​​the rectangle enclosed by the horizontal line where the inductance peak value is located within a fixed period and the preset baseline. This represents the number of sampling points within one period; Indicates the index number of the sampling point; This represents the online estimate of the magnetizing inductance at the i-th sampling point; Indicates the baseline; Indicates the time step; Indicates the peak value of the magnetizing inductance; Indicates the length of a period; This represents the area coefficient.

7. A system for identifying inrush current and fault current of a doubly-fed wind farm's transmitting transformer, characterized in that, The method for identifying inrush current and fault current in a doubly-fed wind farm transmitting transformer as described in any one of claims 1 to 6 includes: A data acquisition model is used to acquire the phase currents on the star side and the phase currents on the delta side of the transformer. The current calculation module is used to convert the phase currents on the star side and the phase currents on the delta side of the transformer to the same reference side, and calculate the differential current. The expression for the differential current is as follows: ； in, Indicates differential current; This represents the current after conversion on the star-connected side of the transformer; This represents the current after delta conversion on the transformer side; The fault identification module is used to compare the differential current with a preset protection starting element setting value. If the differential current is not greater than the preset protection starting element setting value, it is determined to be normal operation or not reached. If the differential current is greater than the preset protection starting element setting value, the inrush current and fault current identification process is initiated. The inrush current and fault current identification process includes: Obtain the phase voltages on the star side of the transformer ; The equivalent circuit of a three-phase transformer is constructed based on the star-delta connection, and the three-phase circuit equation is derived using the phase voltages on the star side and the phase currents on the delta side of the transformer. The three-phase circuit equations are simplified and discretized to obtain the three-phase magnetizing inductance of the transformer. The three-phase magnetizing inductance of the transformer is estimated online using the adaptive forgetting factor least squares method to obtain the online estimated value of the magnetizing inductance at each sampling time. Construct an excitation inductance waveform curve with time as the x-axis and the online estimated value of the excitation inductance as the y-axis. ; Calculate the waveform curve of the excitation inductor The area enclosed by the preset baseline and peak inductance within a fixed period The area of ​​the rectangle enclosed by the horizontal line and the preset baseline. ; Calculate the area and the area of ​​a rectangle The ratio of the two values ​​yields the area coefficient. and the area coefficient With preset protection activation threshold Comparison, if the area coefficient Less than the preset protection activation threshold If the current flowing through the transformer is determined to be an inrush current, the transformer protection is locked out; if the area coefficient is... Not less than the preset protection start threshold If so, the current flowing through the transformer is determined to be a fault current, and an asynchronous method is used to further distinguish the fault type; wherein, the asynchronous method for further distinguishing the fault type includes: The fault type is distinguished by comparing the timing relationship between the differential current surge and the voltage drop. If the differential current surge lags behind the voltage drop, it is judged as an external fault. If the sudden increase in differential current occurs simultaneously with the sudden drop in voltage, then it is determined that a fault has occurred inside the transformer.

Images