A longitudinal protection method and system for offshore wind power low-frequency cable transmission line

Abstract

This invention discloses a longitudinal protection method for low-frequency cable transmission lines of offshore wind power, belonging to the field of power system relay protection technology. Based on the acquired voltage information on both sides of the low-frequency cable line, the voltage surge is determined. When the voltage surge exceeds the preset rated voltage, protection is activated to determine the fault state. The probability distribution of the current on both sides of the low-frequency cable line is determined by the maximum and minimum values ​​of the current within the same data window, yielding a divergence value K. The divergence value K is compared with a setting threshold to determine the type of internal or external fault in the low-frequency cable line. Based on the fault determination result, the corresponding protection operation is executed. This method improves the accuracy of the determination result and enables rapid execution of the corresponding protection operation by extracting the difference in the probability distribution of the current on both sides of the low-frequency cable line, calculating its K-L divergence value K, and constructing a protection criterion for internal and external fault determination.

Description

Technical Field

[0001] This invention relates to the field of power system relay protection technology, and more specifically to a longitudinal protection method and system for offshore wind power low-frequency cable transmission lines. Background Technology

[0002] my country boasts abundant wind resources with broad development prospects. By the end of 2023, the installed capacity of offshore wind power had reached 36.5 million kW. Among the methods of transmitting offshore wind power, Low Frequency Transmission System (LFTS) has become a highly promising grid connection method for transmitting offshore wind power over medium and long distances due to its high transmission capacity and low cost. Relay protection, as the first line of defense for the safe operation of the power system, plays a decisive role in the stable operation of the system.

[0003] In existing relay protection technologies, a ratio-based differential protection method is commonly used. The specific process involves: real-time acquisition of low-frequency current signals from both sides of the transmission line using current transformers installed on both sides. The relative magnitude of the currents on both sides (ratio braking characteristic) is used to determine whether a fault has occurred on the line. When a fault occurs within the line, the difference in current between the two sides increases significantly, exceeding a preset braking threshold. At this point, the protection device will determine it as an in-zone fault. If the determination is an in-zone fault, the protection device will quickly issue a trip signal to disconnect the faulty line.

[0004] In summary, low-frequency transmission systems utilize power electronic components on both sides. Offshore wind farms typically employ permanent magnet synchronous generators (PMSGs), while the grid-connected side of low-frequency cable transmission lines often uses modular multilevel matrix converters (M3Cs) as converters. Once a fault occurs in the low-frequency cable, due to the influence of the power electronic components on both sides, the system's electrical quantities exhibit capacitive characteristics characterized by limited amplitude, controlled phase angle, nonlinearity, and frequency deviation. This leads to a relative increase in current harmonic content, affecting the current measurement results in traditional ratio-based differential protection. This results in increased measurement errors, reduced accuracy of judgment results, and compromised accuracy in the protection device's actions during fault occurrence. Summary of the Invention

[0005] To address the problems existing in the aforementioned fields, this invention proposes a longitudinal protection method and system for low-frequency cable transmission lines of offshore wind power. This method collects voltage information at both ends of the low-frequency cable line, uses the voltage fault component as the protection activation criterion, and collects current information as the raw information for the fault criterion. By extracting the probability distribution of the current at both ends of the low-frequency cable line, the divergence value K is studied when faults occur inside or outside the fault zone of the submarine cable line. The difference in divergence value K is used to distinguish between faults inside and outside the fault zone. This protection method can improve the accuracy of fault determination results, and thus enable targeted and rapid execution of protection based on the determination results.

[0006] To address the aforementioned technical problems, this invention discloses a longitudinal protection method for low-frequency transmission lines of offshore wind power cables, comprising the following steps:

[0007] Obtain current and voltage information from both sides of the low-frequency cable line;

[0008] Based on the voltage information on both sides of the low-frequency cable line, when the voltage surge exceeds the preset rated voltage, the protection is activated to determine the fault status.

[0009] Based on the current information on both sides of the low-frequency cable line, by determining the current direction, the data sampling sequences of the current on both sides of the low-frequency cable line during the half-cycle are placed in the same data window to determine the maximum and minimum values ​​of the current information on both sides of the low-frequency cable line under the same data window; based on the maximum and minimum values ​​of the current, the current data is divided into multiple sub-intervals; by calculating the number of current data on both sides in each sub-interval, the probability distribution of the current on both sides of the low-frequency cable line is determined; based on the probability distribution of the current on both sides of the low-frequency cable line, the divergence value K of the probability distribution is obtained.

[0010] The divergence value K of the probability distribution is compared with the setting threshold. When the divergence value K is higher than the setting threshold, the low-frequency cable line experiences an in-zone fault; when the divergence value K is lower than the setting threshold, the low-frequency cable line experiences an out-of-zone fault.

[0011] Based on the fault diagnosis results, execute the corresponding protection operation.

[0012] Preferably, the fault status determination specifically includes:

[0013] By collecting voltage signals from both sides of the low-frequency cable line, the voltage change is determined, and the voltage change is used as the criterion for protection activation.

[0014] The start-up judgment formula is:

[0015]

[0016] In the formula, voltage change amount This is the voltage at the current time minus the voltage of one wavefront.t This represents the sampled value at the current moment. T The sampling period is The system's rated voltage;

[0017] when If the system malfunctions, the protection mechanism will activate and enter the fault diagnosis state; otherwise, it will not activate.

[0018] Preferably, determining the maximum and minimum values ​​of the current on both sides of the low-frequency cable line under the same data window specifically includes:

[0019] The protection installation point for low-frequency cable lines includes two sampling points to acquire current information from the two sampling points, including the current value of the first sampling point W and the current value of the second sampling point M.

[0020] Taking the direction of current flowing from the busbar to the line as positive, the current value on the W side of the first sampling point is taken as positive, and the current value on the M side of the second sampling point is taken as negative;

[0021] Place the data sampling sequences of the currents on both sides of the first sampling point W and the second sampling point M during the half-cycle into the same data window, and find the maximum value Ma and the minimum value Mi of the data in this window.

[0022] Preferably, determining the probability distribution of the current on both sides of the low-frequency cable line specifically includes:

[0023] Based on the maximum value Ma and the minimum value Mi of the data in this window, the data fluctuation range in this data window is determined to be B = Ma - Mi;

[0024] Divide B into ten sub-intervals, each with a length of 0.1B, numbered from 1 to 10 from low to high. Calculate the number of data points of the current on the first sampling point W side and the second sampling point M side that fall in each sub-interval, and calculate the ratio of the number of data points of the current on the first sampling point W side and the second sampling point M side that fall in the sub-intervals to the total number of sampling points. This is the probability distribution of the current on the first sampling point W side and the second sampling point M side.

[0025] Preferably, obtaining the divergence value K of the probability distribution includes the following steps:

[0026] The divergence value K is calculated using the KL divergence calculation formula.

[0027] The formula for calculating KL divergence is:

[0028]

[0029] in, and The first sampling point W and the second sampling point M are respectively in the calculation data window. iThe probability distribution corresponding to each current data point, where N is the number of data points in the calculation data window. This is a correction factor.

[0030] Preferably, the fault occurring within the low-frequency cable line fault zone specifically includes:

[0031] When a fault occurs within the low-frequency cable zone, the calculated divergence value K satisfies:

[0032]

[0033] In the formula, This is the set threshold;

[0034] The tuning principle is as follows:

[0035]

[0036] In the formula, To consider the reliability coefficient of the phase angle error on both sides of the low-frequency cable line, including the angle error caused by the current transformer and the angle error caused by the cable capacitance current; To account for amplitude errors caused by current transformers; This is the margin coefficient.

[0037] Preferably, the divergence value K for calculating faults outside the occurrence zone of the low-frequency cable line is 0.

[0038] Preferably, it also includes a longitudinal protection system for the low-frequency cable transmission line of offshore wind power, comprising:

[0039] The data acquisition module is used to acquire current and voltage information on both sides of the low-frequency cable line;

[0040] The protection detection module is activated to detect fault conditions when the voltage surge exceeds the preset rated voltage, based on the voltage information on both sides of the low-frequency cable line.

[0041] The fault diagnosis module is used to determine the current direction based on the current information on both sides of the low-frequency cable line. It places the half-cycle current data sampling sequences on both sides of the low-frequency cable line into the same data window, determining the maximum and minimum values ​​of the current information within the same data window. Based on the maximum and minimum values, the current data is divided into multiple sub-intervals. By calculating the number of current data points on both sides within each sub-interval, the probability distribution of the current on both sides of the low-frequency cable line is determined. The divergence value K of the probability distribution is obtained based on this distribution. The divergence value K is compared with a setting threshold. When the divergence value K is higher than the setting threshold, a fault occurs within the fault zone of the low-frequency cable line; when the divergence value K is lower than the setting threshold, a fault occurs outside the fault zone of the low-frequency cable line.

[0042] The protection module is used to perform corresponding protection operations based on the fault diagnosis results.

[0043] Preferably, the device further includes a computer apparatus, the computer apparatus comprising a memory and a processor, the memory storing a computer program, which, when executed by the processor, causes the processor to perform the following steps:

[0044] Obtain current and voltage information from both sides of the low-frequency cable line;

[0045] Based on the voltage information on both sides of the low-frequency cable line, when the voltage surge exceeds the preset rated voltage, the protection is activated to determine the fault status.

[0046] Based on the current information on both sides of the low-frequency cable line, by determining the current direction, the data sampling sequences of the current on both sides of the low-frequency cable line during the half-cycle are placed in the same data window to determine the maximum and minimum values ​​of the current information on both sides of the low-frequency cable line under the same data window; based on the maximum and minimum values ​​of the current, the current data is divided into multiple sub-intervals; by calculating the number of current data on both sides in each sub-interval, the probability distribution of the current on both sides of the low-frequency cable line is determined; based on the probability distribution of the current on both sides of the low-frequency cable line, the divergence value K of the probability distribution is obtained.

[0047] The divergence value K of the probability distribution is compared with the setting threshold. When the divergence value K is higher than the setting threshold, the low-frequency cable line experiences an in-zone fault; when the divergence value K is lower than the setting threshold, the low-frequency cable line experiences an out-of-zone fault.

[0048] Based on the fault diagnosis results, execute the corresponding protection operation.

[0049] Preferably, the system further includes a computer-readable storage medium storing a computer program that, when executed by a processor, causes the processor to perform the following steps:

[0050] Obtain current and voltage information from both sides of the low-frequency cable line;

[0051] Based on the voltage information on both sides of the low-frequency cable line, when the voltage surge exceeds the preset rated voltage, the protection is activated to determine the fault status.

[0052] Based on the current information on both sides of the low-frequency cable line, by determining the current direction, the data sampling sequences of the current on both sides of the low-frequency cable line during the half-cycle are placed in the same data window to determine the maximum and minimum values ​​of the current information on both sides of the low-frequency cable line under the same data window; based on the maximum and minimum values ​​of the current, the current data is divided into multiple sub-intervals; by calculating the number of current data on both sides in each sub-interval, the probability distribution of the current on both sides of the low-frequency cable line is determined; based on the probability distribution of the current on both sides of the low-frequency cable line, the divergence value K of the probability distribution is obtained.

[0053] The divergence value K of the probability distribution is compared with the setting threshold. When the divergence value K is higher than the setting threshold, the low-frequency cable line experiences an in-zone fault; when the divergence value K is lower than the setting threshold, the low-frequency cable line experiences an out-of-zone fault.

[0054] Based on the fault diagnosis results, execute the corresponding protection operation.

[0055] Compared with the prior art, the present invention has the following beneficial effects:

[0056] This invention proposes a longitudinal protection method for low-frequency cable transmission lines in offshore wind power. This method requires rapid fault identification. Due to the limitation of fault current amplitude in dual-sided power electronic sources, this invention uses current and voltage information from both sides of the low-frequency cable line. Based on the voltage information, it determines the voltage surge. When the voltage surge exceeds a preset rated voltage, the protection is activated to determine the fault state. Current information is used as the original information for the judgment. Considering that the protection's speed decreases when the data window is the entire cycle, and the algorithm is affected by fault transients when using a 1 / 4 cycle, severely impacting the calculation results, this invention sets the data window to half a cycle. By determining the current direction, the data sampling sequences of the current on both sides of the low-frequency cable line during the half-cycle are placed in the same data window. The maximum and minimum values ​​of the current on both sides of the low-frequency cable line under the same data window are determined. The current data is divided into multiple sub-intervals. By calculating the number of current data on both sides in each sub-interval, the probability distribution of the current on both sides of the low-frequency cable line is determined. By studying the divergence value K when faults occur inside or outside the fault zone in submarine cable lines, the divergence value K of the probability distribution is compared with the setting threshold to distinguish between faults inside and outside the fault zone, which improves the accuracy of the judgment results. Then, according to different fault types, the corresponding protection operations are quickly executed. Attached Figure Description

[0057] Figure 1 This is a flowchart of the longitudinal protection method for the low-frequency cable transmission line of offshore wind power proposed in this invention.

[0058] Figure 2 This is a structural diagram of a low-frequency offshore wind power transmission system based on M3C, provided for an embodiment of the present invention.

[0059] Figure 3 The current waveform interval division provided for embodiments of the present invention.

[0060] Figure 4 The data provided in each sub-interval during normal operation according to the embodiments of the present invention are as follows: (a) is the data of the first sampling point W distributed in each sub-interval, and (b) is the data of the second sampling point M distributed in each sub-interval.

[0061] Figure 5 Outside the region provided for embodiments of the present invention f Waveform of phase A and K value when there are 5 ABC faults.

[0062] Figure 6 The area provided in the embodiments of the present invention f 3 Waveform of phase A and K value when AG fault occurs.

[0063] Figure 7 The area provided in the embodiments of the present invention f 4 Changes in three-phase waveforms and K value during BC fault.

[0064] Figure 8 The area provided in the embodiments of the present invention f 2 Waveform and K value changes at different transition resistances of AG.

[0065] Figure 9 Outside the region provided for embodiments of the present invention f 1 The waveform and K value changes when BCG data is abnormal.

[0066] Figure 10 The area provided in the embodiments of the present invention f 2 Comparison of protection performance when phase A is grounded through high resistance; where (a) is the performance effect of the corresponding method of the present invention, (b) is the performance effect of the Pearson related method, (c) is the performance effect of the Kenda II related method, and (d) is the performance effect of cosine similarity. Detailed Implementation

[0067] The following will refer to the appendices in the embodiments of the present invention. Figures 1-10 The technical solutions in the embodiments of the present invention will be clearly and completely described. It should be understood that the terminology used in the present invention is only for describing particular implementation methods and is not intended to limit the present invention.

[0068] like Figure 1 As shown, this invention proposes a longitudinal protection method for low-frequency cable transmission lines of offshore wind power, comprising the following steps:

[0069] S1: Obtain the current value at the first sampling point W and the current and voltage values ​​at the second sampling point M of the low-frequency cable.

[0070] S2: Based on the current value at the first sampling point W and the voltage information at the second sampling point M of the low-frequency cable, determine the voltage fluctuation. When the voltage fluctuation exceeds the preset rated voltage, the protection is activated and fault condition judgment begins.

[0071] S3: Based on the current information on both sides of the low-frequency cable line, by determining the current direction, the data sampling sequences of the current on both sides of the low-frequency cable line in the half-cycle are placed in the same data window to determine the maximum and minimum values ​​of the current information on both sides of the low-frequency cable line under the same data window; based on the maximum and minimum values ​​of the current, the current data is divided into multiple sub-intervals; by calculating the number of current data on both sides in each sub-interval, the probability distribution of the current on both sides of the low-frequency cable line is determined; the probability distribution of the current side of the first sampling point W and the probability distribution of the current side of the second sampling point M are substituted into the KL divergence formula to obtain the divergence value K of the probability distribution;

[0072] S4: Compare the divergence value K of the probability distribution with the setting threshold. When the divergence value K is higher than the setting threshold, the low-frequency cable line experiences an in-zone fault; when the divergence value K is lower than the setting threshold, the low-frequency cable line experiences an out-of-zone fault.

[0073] S5: Execute the corresponding protection operation based on the fault diagnosis result.

[0074] In this step, corresponding protection operations need to be performed based on the fault identification results within and outside the zone. The fault status identification specifically includes:

[0075] By collecting voltage signals from both sides of the low-frequency cable line, the voltage change is determined, and the voltage change is used as the criterion for protection activation.

[0076] The start-up judgment formula is:

[0077]

[0078] In the formula, voltage change amount This is the voltage at the current time minus the voltage of one wavefront. t This represents the sampled value at the current moment. T The sampling period is The system's rated voltage;

[0079] When the voltage surge exceeds 0.1 times the rated voltage, that is... If the system malfunctions, the protection mechanism will activate and enter the fault diagnosis state; otherwise, it will not activate.

[0080] Determine the maximum and minimum current values ​​on both sides of the low-frequency cable line within the same data window, specifically including:

[0081] The protection installation point for low-frequency cable lines includes two sampling points to acquire current information from the two sampling points, including the current value of the first sampling point W and the current value of the second sampling point M.

[0082] Taking the direction of current flowing from the busbar to the line as positive, the current value on the first sampling point W side is taken as positive, and the current on the second sampling point M side is taken as negative; the data sampling sequences of the current on both sides of the first sampling point W and the second sampling point M in the half-cycle are placed in the same data window, and the maximum value Ma and minimum value Mi of the current data in the data window are found.

[0083] Determining the probability distribution of current on both sides of a low-frequency cable line specifically includes:

[0084] Based on the maximum value Ma and minimum value Mi of the data in the window, the data fluctuation range in the data window is determined to be B = Ma - Mi; B is divided into ten sub-intervals, each with a length of 0.1B, numbered from 1 to 10 from low to high; the number of data points of the current on the first sampling point W side and the second sampling point M side falling in each sub-interval is calculated respectively, and the ratio of the number of data points of the current on the first sampling point W side and the second sampling point M side falling in the sub-interval to the total number of sampling points is calculated accordingly, which is the probability distribution of the current on the first sampling point W side and the second sampling point M side.

[0085] To obtain the divergence value K of the probability distribution, the following steps are included:

[0086] The divergence value K is calculated using the KL divergence calculation formula.

[0087] The formula for calculating KL divergence is:

[0088]

[0089] in, and The first sampling point W and the second sampling point M are respectively in the calculation data window. i The probability distribution corresponding to each current data point N To calculate the number of data points in the data window, As a correction factor, take 10. -10 .

[0090] Faults occurring within the low-frequency cable line area specifically include:

[0091] When a fault occurs within the low-frequency cable zone, the calculated divergence value K satisfies:

[0092]

[0093] In the formula, This is the set threshold;

[0094] The tuning principle is as follows:

[0095]

[0096] In the formula, To consider the reliability coefficient of the phase angle error on both sides of the low-frequency cable line, including the angle error caused by the current transformer and the angle error caused by the cable capacitance current; To account for amplitude errors caused by current transformers; This is the margin coefficient.

[0097] When the low-frequency cable is operating normally or when an external fault occurs, the calculated divergence value K is 0.

[0098] This invention also proposes a longitudinal protection system for low-frequency cable transmission lines of offshore wind power, comprising:

[0099] The data acquisition module is used to acquire current and voltage information on both sides of the low-frequency cable line;

[0100] The protection detection module is activated to detect fault conditions when the voltage surge exceeds the preset rated voltage, based on the voltage information on both sides of the low-frequency cable line.

[0101] The fault diagnosis module is used to determine the current direction based on the current information on both sides of the low-frequency cable line. It places the half-cycle current data sampling sequences on both sides of the low-frequency cable line into the same data window, determining the maximum and minimum values ​​of the current information within the same data window. Based on the maximum and minimum values, the current data is divided into multiple sub-intervals. By calculating the number of current data points on both sides within each sub-interval, the probability distribution of the current on both sides of the low-frequency cable line is determined. The divergence value K of the probability distribution is obtained based on this distribution. The divergence value K is compared with a setting threshold. When the divergence value K is higher than the setting threshold, a fault occurs within the fault zone of the low-frequency cable line; when the divergence value K is lower than the setting threshold, a fault occurs outside the fault zone of the low-frequency cable line.

[0102] The protection module is used to perform corresponding protection operations based on the fault identification results inside and outside the zone.

[0103] The method proposed in this invention determines the voltage surge based on voltage information from both sides of a low-frequency cable line. When the voltage surge exceeds a preset rated voltage, protection is activated to determine the fault state. The method uses the current sampling values ​​from both sides of the low-frequency cable line as the original information for the criterion. By extracting the difference in the probability distribution of the current on both sides of the low-frequency cable line, the KL divergence value K is calculated. The K value is compared with the setting threshold when a fault occurs inside or outside the fault zone in the low-frequency cable line to construct a protection criterion for fault differentiation between inside and outside the fault zone. This improves the accuracy of the determination results and allows for rapid execution of corresponding protection operations based on different fault connections.

[0104] Example

[0105] To verify the feasibility of the proposed method, the embodiments provided by this invention take a longitudinal protection method for offshore wind power low-frequency cable transmission lines based on KL divergence as an example to verify the longitudinal protection method for offshore wind power low-frequency cable transmission lines proposed by this invention.

[0106] like Figure 2 The diagram shows the structure of the offshore wind power transmission system via low-frequency cable given in this embodiment, including a direct-drive wind turbine generator, a wind turbine converter, a step-up transformer (35 / 220 kV), an isolation transformer, a low-frequency cable, and an M3C converter.

[0107] The generator outlet uses a back-to-back full-power converter to convert wind power into 50 / 3Hz AC power. This AC power is then transmitted via a step-up transformer and a low-frequency cable to the low-frequency side of the M3C converter. The M3C converter then converts the low-frequency AC power into industrial frequency power and integrates it into the onshore power grid. The low-frequency cable includes two sampling points: the first sampling point W and the second sampling point M.

[0108] KL divergence, also known as relative entropy, is a well-known similarity measure in the field of information analysis. It calculates the difference between the information entropy of two probability distributions. For two sets of probability distributions... u and v ,pass u express v The KL divergence value K is:

[0109]

[0110] The discrete form is:

[0111]

[0112] KL divergence is used to measure the distance between two probability distributions. The closer the probability distributions are, the smaller the calculated K value; when the probability distributions are exactly the same, the calculated K value is 0.

[0113] Since KL divergence is often used in the field of information analysis, its input must be a probability distribution, and the data needs to be preprocessed before calculation.

[0114] Considering that the protection speed decreases when the data window takes the entire cycle, and the algorithm is affected by the transient state of the fault when taking 1 / 4 cycle, which seriously affects the calculation results, this invention sets the KL divergence calculation data window to half a cycle (30ms), and calculates the K value sequentially as the sliding data window advances to form a sequence.

[0115] For current probability distribution processing methods such as Figure 3 As shown, taking the current direction from the busbar to the line as positive, firstly... Figure 2 Current in To negate the current, place the data sampling sequences of the currents on both sides during the half-cycle in the same data window, such as... Figure 3 As shown, the maximum value Ma and minimum value Mi of the data in this window are found. Therefore, the data fluctuation range in this data window is B = Ma - Mi. Divide B into ten equal sub-intervals, each with a length of 0.1B, numbered from 1 to 10 from low to high; calculate... and Count the number of data points falling into each sub-interval, and calculate accordingly. and The ratio of the number of data points falling within a sub-interval to the total number of sampling points is the probability distribution, such as... Figure 3 As shown, this illustrates a diagram of data probabilistics.

[0116] Following the above processing method, the number of data points in each sub-interval obtained after the above processing, when the current on both sides of the low-frequency cable transmission line is under the same data window during normal operation, is as follows: Figure 4 As shown.

[0117] Depend on Figure 4 As can be seen, when the system is running normally, the number of data points falling into each sub-interval is basically the same, and even if the number differs, the difference is very small. The calculated data probability distribution is shown in Table 1.

[0118] Table 1. Data probability distribution during normal operation

[0119]

[0120] in, and The first sampling point W and the second sampling point M are respectively in the calculation data window. i The probability distribution corresponding to each interval; to avoid algorithm errors caused by the current sampling probability distribution being 0 in a certain interval, the discrete form of the KL divergence value K is corrected by introducing a correction factor:

[0121]

[0122] In the formula, As a correction factor, this invention uses a probability distribution for calculation. Since it is relatively small, a correction factor is used. .

[0123] By using the discrete form formula of the corrected KL divergence value K, the K value of the data in Table 1 was calculated, and K=0.02 was obtained, which is close to 0. This intuitively reflects that the currents on both sides of the system are very similar when the system is running normally.

[0124] In summary, when the system is operating normally, the current amplitudes on both sides are equal, and the phase angle difference after processing the current on side M is 0, resulting in a very small calculated K value. However, when a fault occurs in the low-frequency cable, the currents on both sides differ in amplitude and phase angle, and the K value is no longer close to 0.

[0125] To meet the requirements of relay protection devices, the present invention constructs the following criteria:

[0126] (1) Protection activation criteria

[0127] Longitudinal protection requires rapid fault identification. Due to the limitation of fault current amplitude in dual-sided power electronic sources, this invention adopts voltage surge as the protection activation criterion:

[0128]

[0129] In the formula, , t This represents the sampled value at the current moment. T The sampling period in this invention is 60ms. The system's rated voltage. When If the system malfunctions, the protection mechanism will activate and enter the fault diagnosis state; otherwise, it will not activate.

[0130] (2) Criteria for Protective Actions

[0131] As the above analysis shows, when the system is operating normally, the calculated K value should be 0. However, when a fault occurs within the zone, the fault currents on both sides will show different differences in amplitude and phase angle. In this case, the calculated K >> 0. Therefore, it is necessary to set an appropriate setting threshold to effectively determine whether a fault has occurred in the low-frequency cable. When a fault occurs within the zone in the low-frequency cable, the calculated K value should satisfy:

[0132]

[0133] In the formula, This is the set threshold.

[0134] Since the KL divergence can simultaneously reflect the difference in amplitude and phase angle of the currents on both sides, The tuning principle is as follows:

[0135]

[0136] In the formula, To account for the reliability coefficient of the phase angle error on both sides of a low-frequency cable line, it generally includes the angle error caused by the current transformer (CT) and the angle error caused by the cable capacitance current. To account for amplitude errors caused by CT; This is the margin coefficient.

[0137] In power systems, the maximum considered phase angle error of current transformers is 8°. The phase angle deviation caused by capacitive current in long lines is generally 7.4° / 100km. The cable length used in this invention is 100km, therefore the phase angle error of the current on both sides of the line is 15.4°, calculated as follows. =0.15; For amplitude variation error in CT measurements, it generally does not exceed 10% on one side. Considering the most severe case, a total of 20% transmission error occurs on both sides. In this case, the calculated... =0.61; As a margin factor, this invention takes 0.4; in summary, =1.2. The protection process of the method provided in this embodiment is as follows: Figure 1 As shown.

[0138] This embodiment uses a longitudinal protection method based on KL divergence for offshore wind power low-frequency cable transmission systems to simulate and analyze low-frequency transmission systems in PSCAD. The low-frequency simulation system and fault point settings are as follows: Figure 2 As shown, the parameters of the system model are as follows:

[0139] On the wind farm side: 50 direct-drive PMSG units, each with a rated capacity of 2MVA, converter port voltage of 0.69kV, and a system base power of 100MW; low-frequency cable outgoing line: 100km long, voltage level of 220kV, transmission frequency of 50 / 3Hz, resistance per unit length of 0.1213Ω / km, inductance of 0.254mH / km, capacitance of 0.1716 / km, and parallel high-voltage reactance of 2×30Mvar. On the M3C side: port voltage of 220kV, port capacitance of 0.5uF. The system sampling frequency is 4kHz, the data window used for calculation is 30ms, and the fault occurrence time is 1.2s. To reduce the influence of high-order harmonics during external faults, this invention performs low-pass filtering on the acquired data.

[0140] Simulations were conducted to verify the performance of various protection systems under different fault conditions (location, type, transition resistance, noise, interference) and high-resistance grounding in low-frequency transmission lines. Please refer to the simulation results for details. Figure 5 , Figure 6 , Figure 7 , Figure 8 , Figure 9 and Figure 10 And Tables 2 and 3.

[0141] in, Figure 5 The waveform of phase A and the K value are given when an ABC fault occurs outside the zone, and the fault occurs at 1.2s. Figure 6 The waveform of phase A and the value of K when an AG fault occurs in the area; Figure 7 The three-phase waveforms and K value during a BC fault within the zone; Figure 8 and Figure 9 The effects of transition resistance and abnormal data on the protection were verified separately. Figure 10 This invention provides a comparison of the protection principles proposed in this invention with those of similar protection principles.

[0142] In practical engineering, transition resistance can also affect the performance of protection systems. The transition resistance of a 220kV transmission line during a ground fault typically does not exceed 100Ω. Figure 8 The protection performance is given when phase A is grounded through different transition resistors within the zone.

[0143] Depend on Figure 8 It can be seen that as the transition resistance increases, the phase angle and amplitude difference of the current waveforms on both sides gradually decrease. However, the protection proposed in this paper focuses on analyzing the relative relationship between the currents on both sides, and is minimally affected by the transition resistance. Even if a ground fault occurs through a 300Ω transition resistance, the protection can still reliably diagnose the fault, demonstrating good resistance to transition resistance.

[0144] During the sampling process of a current transformer, it may be subject to external interference, which may cause errors or asynchrony in the sampled values, resulting in abnormal waveforms in the data window and thus misjudgment. Figure 9 The change in K value is given under the following conditions: during normal operation, a distorted data point appears randomly every 1 / 4 cycle of the current on both sides; and a 1ms error in the two-end communication results in a 10% transmission error.

[0145] Depend on Figure 9 As can be seen, although four distorted data points are generated within each calculation data window, causing changes in the data proportion of some sub-intervals, the impact on the K value remains relatively small. In the most extreme case, all four distorted points fall within the same sub-interval, in which case the probability within that interval will only increase by 3%, and the changes in other sub-intervals will not be significant. While the K value increases slightly when a 1ms error and a transmission error occur in two-way communication, it is still far less than the setpoint.

[0146] Table 2 K values ​​at different fault locations

[0147]

[0148] As shown in Table 2, when a fault occurs within the protection zone of the low-frequency cable transmission line, the K value of the faulty phase is much greater than the setting threshold, and the protection can reliably clear the fault. Meanwhile, the K value of the healthy phase is always less than 1.2, and the protection will not malfunction. When a fault occurs outside the protection zone, the calculated results for all phases are less than the setting threshold, proving the reliability of the protection. Furthermore, Table 2 also verifies that the protection has good fast-acting performance, ensuring fault identification within the second half-cycle of the fault.

[0149] Table 3. K values ​​under different signal-to-noise ratios

[0150]

[0151] As shown in Table 3, although some harmonics are generated after noise is added, the overall trend of the current on both sides is similar, the probability distribution remains almost unchanged, the calculated K value is relatively stable, and the impact of noise on the protection performance is very small.

[0152] according to Figures 5 to 10 The verification results in Tables 2 and 3 clearly demonstrate that the method of the present invention has high sensitivity, good selectivity, fast action speed, and high reliability in detecting faults in low-frequency cable transmission lines of offshore wind power, thereby providing reliable relay protection for transmission lines in offshore wind power low-frequency transmission systems.

[0153] In summary, this invention uses the current sampling values ​​on both sides of a low-frequency cable as the original information for the criterion. By extracting the difference in the probability distribution of the current on both sides of the low-frequency cable line, the KL divergence value K is calculated. Based on the different values ​​of K when the fault occurs inside or outside the submarine cable, a protection criterion is constructed to distinguish between faults inside and outside the area.

[0154] This invention is based on the difference in current at both ends of a low-frequency cable when different faults occur. It analyzes the corresponding fault characteristics, proposes a KL-based criterion, and constructs a relay protection system that is easy to implement and set, with good selectivity and high sensitivity.

[0155] Compared with existing methods, the method proposed in this invention has good speed and reliability, and can more easily and quickly execute protection actions when a fault occurs.

[0156] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

[0157] Furthermore, unless otherwise stated, all technical and scientific terms used in this invention have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. All references to this specification are incorporated by way of citation to disclose and describe methods relating to those references. In the event of any conflict with any incorporated reference, the content of this specification shall prevail.

Claims

1. A method for longitudinal protection of a low frequency cable export line for offshore wind power, c h a r a c t e r i s e d in that, Includes the following steps: Obtain current and voltage information from both sides of the low-frequency cable line; Based on the voltage information on both sides of the low-frequency cable line, when the voltage surge exceeds the preset rated voltage, the protection is activated to determine the fault status. Based on the current information on both sides of the low-frequency cable line, by determining the current direction, the data sampling sequences of the current on both sides of the low-frequency cable line during the half-cycle are placed in the same data window to determine the maximum and minimum values ​​of the current information on both sides of the low-frequency cable line under the same data window; based on the maximum and minimum values ​​of the current, the current data is divided into multiple sub-intervals; by calculating the number of current data on both sides in each sub-interval, the probability distribution of the current on both sides of the low-frequency cable line is determined; based on the probability distribution of the current on both sides of the low-frequency cable line, the divergence value K of the probability distribution is obtained. The divergence value K of the probability distribution is compared with the setting threshold. When the divergence value K is higher than the setting threshold, the low-frequency cable line experiences an in-zone fault; when the divergence value K is lower than the setting threshold, the low-frequency cable line experiences an out-of-zone fault. Based on the fault diagnosis results, execute the corresponding protection operation; Specifically, determining the probability distribution of the current on both sides of the low-frequency cable line includes: Based on the maximum value Ma and minimum value Mi of the current in the data window, the data fluctuation range in the data window is determined to be B = Ma - Mi; Divide B into ten sub-intervals, each with a length of 0.1B, numbered from 1 to 10 from low to high; calculate the number of data points of the current on the first sampling point W side and the second sampling point M side that fall in each sub-interval, and calculate the ratio of the number of data points of the current on the first sampling point W side and the second sampling point M side that fall in the sub-intervals to the total number of sampling points, which is the probability distribution of the current on the first sampling point W side and the second sampling point M side; Obtaining the divergence value K of the probability distribution includes the following steps: The divergence value K is calculated using the KL divergence calculation formula. The formula for calculating KL divergence is: in, and The first sampling point W and the second sampling point M are respectively in the calculation data window. i The probability distribution corresponding to each current data point, where N is the number of data points in the calculation data window. This is a correction factor.

2. The longitudinal protection method for offshore wind power low-frequency cable transmission lines according to claim 1, characterized in that, The fault status determination specifically includes: By collecting voltage signals from both sides of the low-frequency cable line, the voltage change is determined, and the voltage change is used as the criterion for protection activation. The start-up judgment formula is: In the formula, voltage change amount This is the voltage at the current time minus the voltage of one wavefront. t This represents the sampled value at the current moment. T The sampling period is The system's rated voltage; when If the system malfunctions, the protection mechanism will activate and enter the fault diagnosis state; otherwise, it will not activate.

3. The longitudinal protection method for offshore wind power low-frequency cable transmission lines according to claim 1, characterized in that, The determination of the maximum and minimum values ​​of the current on both sides of the low-frequency cable line under the same data window specifically includes: The protection installation point for low-frequency cable lines includes two sampling points to acquire current information from the two sampling points, including the current value of the first sampling point W and the current value of the second sampling point M. Taking the direction of current flowing from the busbar to the line as positive, the current value on the W side of the first sampling point is taken as positive, and the current value on the M side of the second sampling point is taken as negative; Place the data sampling sequences of the currents on both sides of the first sampling point W and the second sampling point M during the half-cycle into the same data window, and find the maximum value Ma and the minimum value Mi of the data window.

4. The longitudinal protection method for offshore wind power low-frequency cable transmission lines according to claim 1, characterized in that, The faults occurring within the low-frequency cable line area specifically include: When a fault occurs within the low-frequency cable zone, the calculated divergence value K satisfies: In the formula, This is the set threshold; The tuning principle is as follows: In the formula, To consider the reliability coefficient of the phase angle error on both sides of the low-frequency cable line, including the angle error caused by the current transformer and the angle error caused by the cable capacitance current; To account for amplitude errors caused by current transformers; This is the margin coefficient.

5. The longitudinal protection method for offshore wind power low-frequency cable transmission lines according to claim 1, characterized in that, The divergence value K for calculating faults outside the occurrence zone of the low-frequency cable line is 0.

6. A longitudinal protection system for low-frequency cable transmission lines of offshore wind power, characterized in that, include: The data acquisition module is used to acquire current and voltage information on both sides of the low-frequency cable line; The protection detection module is activated to detect fault conditions when the voltage surge exceeds the preset rated voltage, based on the voltage information on both sides of the low-frequency cable line. The fault diagnosis module is used to determine the current direction based on the current information on both sides of the low-frequency cable line. It places the half-cycle current data sampling sequences on both sides of the low-frequency cable line into the same data window, determining the maximum and minimum values ​​of the current information within the same data window. Based on the maximum and minimum values, the current data is divided into multiple sub-intervals. By calculating the number of current data points on both sides within each sub-interval, the probability distribution of the current on both sides of the low-frequency cable line is determined. The divergence value K of the probability distribution is obtained based on this distribution. The divergence value K is compared with a setting threshold. When the divergence value K is higher than the setting threshold, a fault occurs within the fault zone of the low-frequency cable line; when the divergence value K is lower than the setting threshold, a fault occurs outside the fault zone of the low-frequency cable line. The protection module is used to perform corresponding protection operations based on the fault diagnosis results; Specifically, determining the probability distribution of the current on both sides of the low-frequency cable line includes: Based on the maximum value Ma and minimum value Mi of the current in the data window, the data fluctuation range in the data window is determined to be B = Ma - Mi; Divide B into ten sub-intervals, each with a length of 0.1B, numbered from 1 to 10 from low to high; calculate the number of data points of the current on the first sampling point W side and the second sampling point M side that fall in each sub-interval, and calculate the ratio of the number of data points of the current on the first sampling point W side and the second sampling point M side that fall in the sub-intervals to the total number of sampling points, which is the probability distribution of the current on the first sampling point W side and the second sampling point M side; Obtaining the divergence value K of the probability distribution includes the following steps: The divergence value K is calculated using the KL divergence calculation formula. The formula for calculating KL divergence is: in, and The first sampling point W and the second sampling point M are respectively in the calculation data window. i The probability distribution corresponding to each current data point, where N is the number of data points in the calculation data window. This is a correction factor.

7. A computer device, characterized in that, The computer device includes a memory and a processor, the memory storing a computer program that, when executed by the processor, causes the processor to perform the method as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, causes the processor to perform the method as described in any one of claims 1 to 5.

Images