New energy station sending-out line pilot protection method, system and device

Abstract

The invention discloses a pilot protection method, system and device for a new energy station sending-out line, and relates to the technical field of new energy station sending-out line protection.The method comprises the steps that transient fault current signals of the new energy station side and the system side on the two sides of a fault are decomposed into different frequency components through DCT; obtaining a two-dimensional DCT coefficient corresponding to the current data at each time point, and generating a binary image; the binary images are flattened into one-dimensional vectors to obtain respective perceptual hash values, the Hamming distance between the two perceptual hash values is calculated and compared to judge the similarity of waveforms, and pilot protection is carried out according to the similarity; according to the method, fault positioning of a sending-out line and accurate action of a protection device are achieved, the method is suitable for a network-forming type large-scale new energy sending-out system, and rapid and efficient fault processing is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of new energy transmission line protection, and in particular to a pilot protection method, system and device for a new energy station transmission line. Background Art

[0002] For a long time, China has constructed a number of large-scale AC synchronous power grids dominated by thermal power, hydropower, etc. With the promotion of the dual-carbon goal, the proportion of new energy power generation will continue to increase. However, there are still various problems in the development of new energy, such as power and electricity balance problems, transient stability problems, etc. The grid-forming technology can improve the system inertia and short-circuit capacity, increase the short-circuit ratio of multiple new energy stations, and improve the grid impedance characteristics, which can effectively reconcile various problems faced by the development of new energy. It is one of the key technologies for the construction of future new power systems, and its importance for the research on the relay protection of the converter AC system of the new energy transmission system is self-evident.

[0003] At present, relay protection devices based on power frequency quantities designed specifically for synchronous machine systems are still widely used for the outgoing lines of new energy AC power grids. However, the external characteristics of grid-forming converters and grid-following converters with current source characteristics are quite different, and there are certain differences in the dynamic responses during the fault ride-through period of traditional synchronous machine systems. When a fault occurs in the new energy converter AC system, the fault characteristics of the system will change significantly, resulting in a significant decline in the operating performance of traditional power frequency quantity-based protection methods, and even misoperation of the protection may occur. For example, the literature "AC Line Protection for Wind Power Transmitted via VSC-HVDC Based on Matrix Mutation Characteristics" proposes a protection principle based on current quantity. It can more sensitively reflect the amplitude difference degree of current mutation using the matrix mutation method. However, the selection of its setting value depends on simulation tests and on-site operation records, and it is easily affected by transition resistance and time window length; the literature "Pilot Protection for New Energy Station Outgoing Lines Based on Differential Current Energy Ratio" proposes a pilot protection criterion based on differential current energy ratio. However, this protection is easily affected by transition resistance and noise; the literature "Application of Tellegen's Apparent Power Protection in Double-Circuit AC Transmission Lines on the Same Tower" proposes a protection method based on power quantity according to the law of conservation of power. However, this type of method is easily affected by the voltage dead zone caused by metal short-circuit faults at the protection outlet, and has high requirements for the configuration of communication devices for electrical quantity data at both ends of the transmission.

[0004] To sum up, the currently proposed pilot protection methods are easily affected by transition resistance, time window length, and the voltage dead zone caused by metal short-circuit faults at the protection outlet, resulting in errors in the fault location of the new energy station outgoing line and unable to achieve accurate operation of the discriminant protection device. Summary of the Invention

[0005] In view of the deficiencies of the existing technology, which is vulnerable to the influence of transition resistance, time window length, and voltage dead zones caused by metallic short - circuit faults at the protection outlet, resulting in errors in the fault location of the outgoing lines of new - energy power stations, the present invention proposes a pilot protection method, system, and device for the outgoing lines of new - energy power stations, thereby solving the problems existing in the existing technology.

[0006] A pilot protection method for the outgoing lines of new - energy power stations includes the following steps: Collect the current data on the new - energy power station side and the grid side within one cycle after a fault occurs in the outgoing lines of the new - energy power station; Decompose the current data on the new - energy power station side and the grid side into different frequency components through DCT transform to obtain the two - dimensional DCT coefficients corresponding to the current data at each time point, and convert the two - dimensional DCT coefficients into binary images; flatten the binary images of the current data on the new - energy power station side and the grid side into one - dimensional vectors, and use them as the respective corresponding perceptual hash values; calculate the respective corresponding Hamming distances according to each perceptual hash value; Perform pilot protection by comparing the Hamming distances of the two perceptual hash values; among them, if the Hamming distance is greater than the set threshold of the in - zone and out - of - zone fault criterion, it is determined that an in - zone fault has occurred in the corresponding phases on the new - energy power station side and the grid side, and protection actions are taken; otherwise, it is determined as an out - of - zone fault, and protection restoration is performed.

[0007] Further, the process of decomposing the current data on the new - energy power station side and the grid side into different frequency components through DCT transform to obtain the two - dimensional DCT coefficients corresponding to the current data at each time point is specifically expressed as: ; Among them, is the sampled current data, X k is the two - dimensional DCT coefficient, N is the total number of sampling points of the current data within one cycle, n represents the sequence number of the collected current data, k is the frequency index.

[0008] Further, the process of converting the two - dimensional DCT coefficients into binary images specifically includes the following steps: Select the first 64 DCT coefficients and calculate their mean value, and determine the threshold according to the mean value; Compare the DCT coefficients with the threshold. If it is greater than or equal to the threshold, the DCT coefficient is represented as 1. If it is less than the threshold, the DCT coefficient is represented as 0, thereby generating a binary image.

[0009] Further, the Hamming distance calculated according to the perceptual hash value is represented as: ; Among them, h 1 and h 2 are binary hash values, and ⊕ represents a bitwise exclusive OR operation; the Hamming distance is obtained by calculating the number of 1s in the exclusive OR result.

[0010] The present invention also includes a pilot protection system for the outgoing line of a new energy power station, including: An acquisition module, configured to acquire the current data on the new energy power station side and the grid side within one cycle after a fault occurs in the outgoing line of the new energy power station; A calculation module, configured to decompose the current data on the new energy power station side and the grid side into different frequency components through DCT transformation, obtain the two-dimensional DCT coefficients corresponding to the current data at each time point, and convert the two-dimensional DCT coefficients into binary images; flatten the binary images of the current data on the new energy power station side and the grid side into one-dimensional vectors, and use them as the respective corresponding perceptual hash values; calculate the respective corresponding Hamming distances according to each perceptual hash value; A determination module, configured to perform pilot protection by comparing the Hamming distances of two perceptual hash values; among them, if the Hamming distance is greater than the set threshold of the in-zone and out-of-zone fault criterion, it is determined that an in-zone fault has occurred in the corresponding phase on the new energy power station side and the grid side, and a protection action is performed; otherwise, it is determined as an out-of-zone fault, and a protection reset is performed.

[0011] The present invention also includes a computer device for pilot protection of the outgoing line of a new energy power station, including: a memory, a processor, and a computer program stored in the memory, and when the processor executes the computer program, the steps of the pilot protection method for the outgoing line of the new energy power station are implemented.

[0012] The present invention also includes a readable storage medium, the readable storage medium stores a computer program, the computer program includes program instructions, and when the program instructions are executed by a processor, the steps of the pilot protection method for the outgoing line of the new energy power station are executed.

[0013] The present invention provides a pilot protection method for the outgoing line of a new energy power station, having the following beneficial effects: The present invention directly cuts into from the image recognition link of the current waveform itself, and the DCT transformation link effectively decomposes current components of different frequencies, retains and amplifies effective low-frequency components, and filters out abnormal high-frequency components, greatly reducing the adverse effects of the control link on line protection; at the same time, the similarity of waveforms is judged by calculating and comparing the Hamming distances of two perceptual hash values, realizing the fault location of the outgoing line and the accurate operation of the protection device. The principle is simple, not affected by the fault location, fault type, and transition resistance, and is applicable to a large-scale new energy outgoing system with a network structure, realizing fast and efficient fault handling. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 Schematic diagram of the network - forming new - energy grid - connection topology and control link in the embodiment of the present invention; Figure 2 Topology diagram of the network - forming new - energy power source grid - connection in the embodiment of the present invention; Figure 3 Phasor diagram of fault currents on both sides of the system during line faults in the embodiment of the present invention; Figure 3 Case a in [[]] means leading and both are in the second or third quadrant of the fault - current phasor diagram; Figure 3 Case b in [[]] means lagging and both are in the first or fourth quadrant of the fault - current phasor diagram; Figure 4 Schematic diagram of the analysis results of the adaptability of traditional current differential protection under two - phase - to - ground short - circuit faults in the embodiment of the present invention; Figure 4 In (a) of [[]] Figure 4 is the analysis diagram of the phase - angle difference of three - phase currents on both sides; Figure 4 In (b) of [[]] is the analysis diagram of the ratio of three - phase current differential to braking current on both sides; Figure 5 Operation flow chart of the pilot protection method for the network - forming new - energy transmission line based on the perceptual hashing algorithm in the embodiment of the present invention; Figure 6 Schematic diagram of the A - phase current waveforms on both sides of the transmission line during an external AG fault in the embodiment of the present invention; Figure 7 Schematic diagram of the A - phase current waveforms on both sides of the transmission line during an external AG fault in the embodiment of the present invention; Figure 8 DCT binary image obtained by DCT transformation of the A - phase current waveforms on both sides of the transmission line during an internal AG fault in the embodiment of the present invention; Figure 8 In (a) of [[]] Figure 8 is the overall DCT binary image of the new - energy side and the system side, Figure 8 In (b) of [[]] is the local DCT binary image corresponding to the first 64 DCT coefficients generated by the new - energy side and the system side. Detailed implementation manners

[0015] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments.

[0016] The present invention proposes a pilot protection method for the outgoing line of a network-forming new energy system. The transient fault current signals on the new energy power station side and the system side of both sides of the fault are decomposed into different frequency components through DCT transformation to calculate their respective two-dimensional DCT coefficients, and a binary image is generated. Finally, by flattening the DCT binary image into a one-dimensional vector, the respective perceptual hash values are obtained. By calculating and comparing the Hamming distance of the two perceptual hash values, the similarity of the waveforms is judged, so as to realize the fault location of the outgoing line and the accurate operation of the protection device. The principle is simple, not affected by the fault location, fault type and transition resistance, and is applicable to the network-forming large-scale new energy outgoing system to realize fast and efficient fault handling.

[0017] 1. Analysis of operating characteristics The network-forming new energy grid connection topology and control links are as Figure 1 shown. Dcontrol and Vcontrol in the inner loop control link represent droop control and virtual synchronous machine control respectively, and a control mode is selected for the system by means of a dial. is the voltage on the DC side of the three-phase inverter, is the DC bus capacitor, is the output current, is the converter output voltage, PCC (Point of Common Coupling, PCC) is the point of common coupling, is the grid connection point voltage, , are the equivalent resistance and equivalent inductance between the converter port and the PCC point respectively, , are the line equivalent resistance and equivalent inductance respectively; P ref and P e are the active reference power and the input electromagnetic power respectively, Q ref and Q e are the reactive reference power and the input reactive electromagnetic power respectively, , and are the angular velocity reference value, the output angular velocity value and the input angular velocity respectively. J and D are the inertia and damping values of the virtual synchronous machine control link respectively, is the output angle value, U o is the input system voltage value, E ref is the output voltage reference value, 1 / s is the operator of the inverse Laplace transform, k is a multiple value and has no physical meaning; Kp is the droop coefficient of the droop control link, △ and △ U are the angular velocity change and voltage change calculated by the active - frequency link and the reactive - voltage link respectively. abc 、 dq0 represent three - phase transformation and dq transformation.

[0018] As Figure 2 shown, the DC power on the new - energy side is inverted into AC power by a grid - forming inverter and then boosted to 220 kV, and then sent to the large - scale power grid through the transmission line.

[0019] 2. Fault feature analysis: From Figure 2 it can be seen that due to the negative - sequence elimination control on the inverter side, the zero - sequence component is isolated when a fault occurs in the transmission line of the grid - forming converter. Then, the fault current provided by the new - energy side only contains the positive - sequence component, and its expression is: (1) Among them, 、 are the direct - axis and quadrature - axis current reference values respectively, is the fault initial phase angle, is the direct - axis current value when no fault occurs, 、 are the damping ratio and damping angle respectively, is the reference frequency, is the damped oscillation frequency.

[0020] Combining Figure 1 and Figure 2 , by establishing the basic circuit equation and each control - link equation, the analytical formula of the short - circuit current considering the double - loop response of voltage and current is derived under symmetrical and asymmetrical fault scenarios: (2) (3) The above two formulas are the transient short - circuit current dq coordinate - system expressions under symmetrical and asymmetrical faults respectively. In the first formula, 、 、 and are the amplitudes and initial phase - angle of the axis components of each current dq axis respectively, and are both decay time constants related to the inverter ~ virtual impedance . In the second formula 、 、 、 and They are all arithmetic expressions composed of factors such as the virtual impedance inductance-resistance ratio, the phase angle of the grid connection point voltage, and other parameters. Since the expression form is complex, it will not be elaborated in detail here.

[0021] Perform dq / abc coordinate transformation on the above expressions to obtain the three-phase short-circuit current expressions. Taking phase A as an example, we have: (4) (5) The above two formulas are the three-phase expressions of the transient short-circuit current under symmetrical faults and asymmetrical faults respectively. In the first formula, 、 are the amplitude coefficients of each component in the three-phase current and the initial phase angles of each AC component under symmetrical faults respectively; similarly, in the second formula, 、 are the amplitude coefficients of each component in the three-phase current and the initial phase angles of each AC component under asymmetrical faults respectively. It can be seen from the above two formulas that regardless of whether it is a symmetrical fault or not, the three-phase short-circuit current is composed of a steady-state power-frequency component and three transient components. Among them, the double-frequency decaying transient component is transformed from the component rotating forward in the dq coordinate system to the three-phase stationary abc coordinate system, and each component is closely related to the virtual impedance link both in amplitude and phase angle.

[0022] In summary, when a fault occurs in the outgoing line, the fault current on the new energy side contains a decaying non-power-frequency sine wave and only contains positive-sequence components, while the fault current on the grid side contains double-frequency components, and the lack of current limiting control causes its fault current amplitude to reach several times or even dozens of times the rated current at most. Therefore, the fault currents on the new energy side and the grid side are quite different.

[0023] 3. Analysis of the adaptability of traditional differential protection: The basic operation criterion of the commonly used full-scale phase current differential protection is as follows: (6) Among them, 、 are the amplitudes of the differential and braking currents of each phase respectively, φ represents the phase; is the action threshold of the differential current, 、 are the phase currents of the same name at the protection installation locations on the new energy side and the grid side of the outgoing line respectively, and the directions are both from the bus to the line.

[0024] Taking the AB-phase short-circuit fault in the area as an example, since the inverter suppresses the negative-sequence component, the initial phase angle difference of the fault phase currents on the new energy side is fixed at 120°, and the characteristics are not affected by the fault conditions. According to Kirchhoff's current law, the total short-circuit fault current of the AB phases on the grid side always satisfies: (7) In the formula, 、 and are the corresponding phase currents at the protection installation points on the grid side and the new energy side respectively. Since the phase currents on the system side always maintain a 120° phase difference between two phases, so ; is the line distributed capacitance current, and the unit of both is kA.

[0025] It can be seen from equations (2) to (5) that the amplitude coefficient of the transient fault current output by the grid-forming converter is related to factors such as the resistance-to-inductance ratio of the virtual impedance, the leading phase angle of the virtual internal potential with respect to the grid-connected point voltage, and the magnitudes of the grid-connected point voltages before and after the fault. Compared with the traditional system, the change of the fault current amplitude is more complex and there are more uncertain factors. In addition, due to the existence of the line distributed capacitance, the phase angle difference between the fault currents of phases A and B on the grid side is no longer approximately 180°. The phasor diagram is drawn with this as the reference benchmark as Figure 3 shown.

[0026] The fault current phasors are discussed according to two cases. As shown in case a in Figure 3 , leads , and both are located in the second or third quadrant. At the same time, since is small, has a smaller phase angle difference compared to . Therefore, also leads . At this time, the phase angle difference between and is between 0° and 180°. Since the phase angle differences between , and are 120° to each other, so is between 120° and 180°, and is between 60° and 180°. When or exceeds 90° in some fault scenarios, the corresponding phase braking current is greater than the differential current, and the sensitivity of the protection for this phase decreases, and the risk of refusal to operate increases greatly.

[0027] As shown in case b in Figure 3 , lags , and both are located in the first or fourth quadrant. At the same time, since Smaller, Compared with the phase angle difference is smaller, so it also lags behind . At this time, both are between 0 and 120°. When in some fault scenarios or exceeds 90°, the corresponding phase restraint current is greater than the differential current, and the sensitivity of the protection for this phase decreases, and it may even refuse to operate.

[0028] To verify the above inference, taking the ABG two-phase grounding short-circuit fault occurring at the sending line F3 as an example, the protection restraint value is taken as 0.8, and the adaptability analysis diagram of the traditional current differential protection is as shown in Figure 2 Figure (a) of Figure 4 shows that the phase angle difference of the current after the A-phase fault is about 50°, while the phase angle difference of the current after the C-phase fault is about 100°; Figure 4 In figure (b) of Figure 4 the ratio of the differential current to the restraint current of the A-phase is significantly greater than the restraint value, and the protection can operate reliably; while the ratio of the differential current to the restraint current of the C-phase is less than the restraint value, and the protection refuses to operate.

[0029] 4. Pilot protection principle The present invention takes the network-constructing type new energy sending system shown in Figure 2 as the analysis model. The new energy power station is located on the M side of the 220 kV sending line and is connected to the AC power grid via the line MN. Five fault points are set in the line, where F1 and F5 are external faults, and F2, F3, and F4 are internal faults.

[0030] Since the positive direction of the current is usually defined as from the bus to the line, after an external fault occurs, if the line capacitance current is ignored, the two-side currents , satisfy , which means and have almost exactly the same waveform. When an internal fault occurs, according to the above-mentioned fault current analysis, the frequency and phase of the fault current on the new energy side are distorted and do not completely present as a power-frequency sine form, and due to the current-limiting control of the virtual impedance, the fault current amplitude on the new energy side is much smaller than the fault current on the power grid side. Therefore, there are significant differences in the waveforms of the fault currents on both sides of the sending line, and this feature can be used to distinguish internal and external faults.

[0031] 5. Perceptual hashing algorithm The perceptual hashing algorithm is a hashing algorithm based on perceptual similarity, mainly used to generate "fingerprints" or "abstracts" of multimedia content such as images. By perceptual features (such as shapes and color distributions in images), it finds similar content, and then judges the similarity between the two by calculating and comparing the hash values. The basic idea of using the perceptual hashing algorithm to measure the similarity of current waveforms on both sides of the outgoing line is as follows: The sampled current data on both sides extracted is decomposed into different frequency components through DCT transformation to calculate the respective two-dimensional DCT coefficients, and a binary image is generated; the DCT binary image is flattened into a one-dimensional vector to obtain the respective perceptual hash values, and the similarity of the waveforms is judged by calculating and comparing the Hamming distance between the two perceptual hash values, so as to construct a protection criterion to distinguish internal faults from external faults and normal operating states.

[0032] 6. Pilot protection criterion: (1) Starting criterion; The starting criterion of the pilot protection is shown as follows: (8) (9) In the formula: i ph is the sampled value of the ph-phase current, i 0 is the zero-sequence current, N is the number of sampling points per cycle, I set is the threshold value of the starting criterion. When the sampled values of any phase current for three consecutive times satisfy any one of the formulas in the starting criterion, the protection starts and continues with the subsequent steps; otherwise, the protection resets.

[0033] (2) Identification criterion for internal and external faults: During normal operation and after an external fault occurs, the currents on both sides of the outgoing line I M , - I N have almost exactly the same waveform images, so the calculated result of the Hamming distance output HD ( h 1, h 2) is zero or a relatively small positive number (with errors considering external interference factors); while when an internal fault occurs, the currents on both sides of the outgoing line I M , - I N have significantly different waveform images, corresponding to HD ( h 1, h 2) with a larger result (not exceeding 64 at most). Therefore, the following criterion can be obtained: (10) In the formula: Hset is the threshold value for the internal and external fault criterion. To avoid the influence of interference factors such as noise, H set it is set as a small positive number. If the calculated Hamming distance result HD ( h 1, h 2) satisfies the internal and external fault criterion as shown above, it is determined that an internal fault occurs in the corresponding phase, a tripping command is issued for the protection action, and the protection process ends; otherwise, it is determined that an external fault occurs and the protection is reset.

[0034] The pilot protection method specifically includes the following steps: S1. Obtain the three-phase current sampling values at both the new energy power station side and the system side of the outgoing line, and collect the current data within one cycle after the fault occurs; specifically: The present invention takes the Figure 2 configured grid-connected new energy outgoing system shown as the analysis model. The new energy power station is located on the M side of the 220 kV outgoing line and is connected to the AC power grid via the line MN. Five fault points are set in the line, where F1 and F5 are external faults, and F2, F3, and F4 are internal faults.

[0035] Since the positive direction of the current is usually defined as from the bus to the line, after an external fault occurs, if the line capacitance current is ignored, the currents on both sides , satisfy , which means and have almost exactly the same waveform. When an internal fault occurs, according to the foregoing fault current analysis, the frequency and phase of the fault current on the new energy side are distorted and do not completely present as a power frequency sine form. Moreover, due to the current limiting control of the virtual impedance, the magnitude of the fault current on the new energy side is much smaller than the fault current on the grid side. Therefore, there are significant differences in the waveforms of the fault currents on both sides of the outgoing line, and this feature can be used to distinguish internal and external faults.

[0036] S2. Pilot protection criterion: (1) The starting criterion is shown as the following formula: (11) (12) In the formula: i ph is the ph-phase current sampling value, i 0 is the zero-sequence current, k represents the k th sampling point, N represents the number of sampling points per cycle,I set It is the threshold value for the starting criterion. When any three consecutive sampling values of the phase current satisfy any one of the starting criteria, the protection starts; otherwise, the protection is reset.

[0037] S3. Decompose the sampled current data on both sides through DCT transform into different frequency components, thereby calculating their respective two-dimensional DCT coefficients and generating a binary image. In many applications (such as image compression, pattern recognition, etc.), the luminance information of the image is more important than the color information. After such transformation, the DCT transform can be concentrated on the analysis of the luminance information, thereby improving the algorithm efficiency and effect; the discrete cosine transform (DCT) is an effective method for image feature extraction. By transforming the image from the spatial domain to the frequency domain, it can separate the low-frequency information (such as the general shape and region) and high-frequency information (such as details and textures) in the image. The DCT transform can effectively extract the main structural features of the time-domain current waveform. The low-frequency components usually carry most of the information of the image, removing the detailed parts; the calculated DCT coefficient matrix represents the intensity of the image at different frequencies. For tasks such as image transformation and compression, the DCT coefficients of the time-domain data are more concise and efficient than other coefficients.

[0038] DCT decomposes the image into different frequency components. The low-frequency components usually contain most of the image information, while the high-frequency components are mainly responsible for details and edges. The definition of the DCT transform is shown in the following formula: (13) In the formula: I n is the original sampled current data, n represents the sequential label of the sampled current data, X k is the transformed frequency-domain signal; N is the total number of signal samples. Since the simulation system and the in-loop test platform adopted collect the current waveform data within a 20-ms data window at a sampling frequency of 10 kHz under the standard power frequency of 50 Hz, there are a total of 200 sampling points, that is N = 200; k is the frequency index. Since the research is carried out under the standard power frequency of 50 Hz, so k takes 1. This formula means converting the time-domain signal into a frequency-domain signal through the cosine function to obtain the two-dimensional DCT coefficients. Select the first 64 DCT coefficients and calculate their mean value to determine a threshold for converting the coefficients into a binary image. Compare the DCT coefficients with the mean value. The coefficients greater than or equal to the mean value are set to 1, and the coefficients less than the mean value are set to 0, thereby generating a binary image.

[0039] S4. After the fault occurs on the outgoing line, select the current data at both ends of the fault point, i.e., the new energy side and the system side, in the first cycle after the fault, which is 20 ms. Flatten the binary images of the two groups of current data into one-dimensional vectors and use them as their respective perceptual hash values. Determine the similarity of the waveforms by calculating and comparing the Hamming distance between the two perceptual hash values. The calculation method of the Hamming distance is as follows: (14) In the formula: h 1 and h 2 are binary hash values, that is, the two perceptual hash values. ⊕ represents the bitwise exclusive OR operation. The number of 1s in the exclusive OR result is calculated as the Hamming distance. Next, the calculation result of the Hamming distance is simply referred to as HD ( h 1, h 2). The criteria for judging internal and external faults of pilot protection are shown in the following formula: (15) In the formula: H set is the threshold value of the criteria for judging internal and external faults. To avoid the influence of interference factors such as noise, H set is set as a small positive number. If the calculated Hamming distance result HD ( h 1, h 2)meets the criteria for judging internal and external faults shown above, it is determined that an internal fault has occurred in the corresponding phase, and a tripping command is issued for the protection action, and the protection process ends; otherwise, it is determined that an external fault has occurred, and the protection is reset.

[0040] Simulation analysis: Build a networked new energy outgoing system as shown in Figure 1 in PSCAD. The outgoing voltage level is 220 kV, the power grid frequency is 50 Hz, the line length is 40 km, the new energy side capacity is 80 MW, and the damping coefficient D p = 100 in the VSG control link, and the inertia coefficient J = 2. The data window is selected as 20 ms, and the sampling frequency is 10 kHz. Considering the reliability and sensitivity of the protection comprehensively, when the value range of the calculation result is [0 64], the error does not exceed 10%. Therefore, the criterion threshold H set is set to 6. A total of five faults are set. F1 is an external fault on the new energy side, F5 is an external fault on the large power grid side, and F2, F3, and F4 are internal faults at 1 km, 20 km, and 39 km from the new energy side on the transmission line respectively. Four fault types are set, including single-phase ground fault AG of phase A, phase AB fault AB, phase AB ground fault ABG, and three-phase fault ABC; three fault transition resistancesR f , including 0.01 Ω, 100 Ω, and 300 Ω. The fault moment is 0 ms. Taking the cases where the AG fault occurs outside and inside the zone as examples, the waveforms of the phase A current at both ends of the line are as Figure 6 , Figure 7 shown.

[0041] From Figure 6 , Figure 7 it can be seen that the waveforms of the fault currents at both ends of the line corresponding to the occurrence of faults inside and outside the zone are significantly different. When the fault occurs outside the zone, the waveform is axisymmetric about the horizontal axis. After the waveform data of the fault currents at both ends of the line are processed by DCT transformation to obtain two-dimensional DCT coefficients, the first 64 DCT coefficients are selected and their mean values are calculated to determine a threshold for converting the coefficients into a binary image. The generated DCT binary image is as Figure 8 shown, Figure 8 where (a) in Figure 8 is the overall DCT binary image of the new energy side and the system side, and (b) in Figure 8 is the local DCT binary image corresponding to the first 64 DCT coefficients generated by the new energy side and the system side. On this basis, the binary images of the two groups of current data are flattened into one-dimensional vectors, and they are used as their respective perceptual hash values. Furthermore, the Hamming distance between the two perceptual hash values is calculated to measure the similarity of the waveforms.

[0042] Taking the single-phase fault as an example, considering that the maximum transition resistance of the 220 kV transmission line is about 100 Ω and the 150 Ω high-resistance grounding fault instances of the 220 kV transmission line that have occurred in the past, set the AG fault to occur at positions F1, F3, and F5, and the transition resistances are 50, 100, and 300 Ω respectively. It can be seen from Table 1 that under different transition resistances, the protection can accurately identify the specific fault phase of the fault inside the zone; for the faults outside the zone and the non-fault phases of the faults inside the zone, the protection can make correct judgments. Therefore, the proposed protection method has extremely strong anti-transition resistance ability.

[0043] Table 1 Simulation results under different transition resistances

[0044] Set the outgoing line to occur single-phase ground fault (AG), two-phase short-circuit fault (AB), two-phase ground short-circuit fault (ABG), and three-phase short-circuit fault (ABC). The simulation results are shown in Table 2. It can be seen from Table 2 that when the fault occurs outside the zone, the calculated HD ( h 1 ，h 2) are all less than the setting value of 6, and the protection reliably does not operate; for various faults occurring inside the zone, the HD ( h 1 ，h2) They are all significantly greater than the setting value of 6, and the protection can sensitively and accurately identify the faulty phase. Therefore, the proposed protection method is not affected by the fault location and fault type.

[0045] Table 2 Simulation results under different fault conditions

[0046] In summary, through theoretical and simulation analysis, it can be seen that the proposed method can effectively cope with different fault types, different fault locations, different transition resistances and other conditions, and can sensitively and reliably identify internal and external faults. Traditional pilot protection methods emphasize strong anti-synchronization error ability. Moreover, when complex control links are adopted at the power generation end, the output current waveform is easily affected by adverse factors such as sharp increase in high-frequency components, waveform distortion and phase shift, resulting in a significant decline in protection performance. Compared with other similarity protection methods, the DCT transformation link can effectively decompose current components of different frequencies, retain and amplify effective low-frequency components, filter out abnormal high-frequency components, and greatly reduce the adverse impact of the control link on line protection; at the same time, it is not affected by the anti-synchronization error ability, has a natural fault phase selection ability, and can operate correctly under fault conditions where the sensitivity of other protection methods decreases or even refuses to operate, with high reliability and sensitivity.

[0047] Based on the same inventive concept, the present invention also proposes a pilot protection system for the outgoing line of a new energy power station, including: An acquisition module, configured to acquire the current data on the new energy power station side and the grid side within one cycle after a fault occurs in the outgoing line of the new energy power station.

[0048] A calculation module, configured to decompose the current data on the new energy power station side and the grid side into different frequency components through DCT transformation, obtain the two-dimensional DCT coefficients corresponding to the current data at each time point, and convert the two-dimensional DCT coefficients into binary images; flatten the binary images of the current data on the new energy power station side and the grid side into one-dimensional vectors, and use them as the respective corresponding perceptual hash values; calculate the respective corresponding Hamming distances according to each perceptual hash value.

[0049] A determination module, configured to perform pilot protection by comparing the Hamming distances of the two perceptual hash values; wherein, if the Hamming distance is greater than the set threshold of the internal and external fault criterion, it is determined that an internal fault occurs in the corresponding phase on the new energy power station side and the grid side, and a protection action is performed; otherwise, it is determined as an external fault, and a protection reset is performed.

[0050] The present invention also proposes a computer device for pilot protection of the outgoing line of a new energy power station, including: a memory, a processor, and a computer program stored in the memory. When the processor executes the computer program, the steps of the pilot protection method for the outgoing line of the new energy power station are implemented.

[0051] The present invention also provides a readable storage medium storing a computer program, where the computer program includes program instructions, and when the program instructions are executed by a processor, they are used to perform the steps of the longitudinal protection method for the outgoing line of a new energy power station.

[0052] As described above, the above are only the preferred specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution and inventive concept of the present invention, making equivalent substitutions or changes should be covered within the protection scope of the present invention.

Claims

1. A pilot protection method for the outgoing line of a new energy power station, characterized in that, It includes the following steps: Collect the current data on the new energy power station side and the grid side within one cycle after the occurrence of a fault in the outgoing line of the new energy power station; Decompose the current data on the new energy power station side and the grid side into different frequency components through DCT transformation to obtain the two-dimensional DCT coefficients corresponding to the current data at each time point, and convert the two-dimensional DCT coefficients into binary images; Flatten the binary images of the current data on the new energy power station side and the grid side into one-dimensional vectors, and use them as the respective corresponding perceptual hash values; Calculate the respective corresponding Hamming distances according to each perceptual hash value; Perform pilot protection by comparing the Hamming distances of the two perceptual hash values; Among them, if the Hamming distance is greater than the set threshold of the in-zone and out-of-zone fault criterion, it is determined that an in-zone fault has occurred in the corresponding phase on the new energy power station side and the grid side, and protection actions are taken; Otherwise, it is determined as an out-of-zone fault and protection restoration is performed.

2. The pilot protection method for the outgoing line of a new energy power station according to claim 1, characterized in that, The step of decomposing the current data on the new energy power station side and the grid side into different frequency components through DCT transformation to obtain the two-dimensional DCT coefficients corresponding to the current data at each time point; The specific process is expressed as: ； Among them, is the sampled current data, X k are two-dimensional DCT coefficients, N is the total number of sampled points of current data within one cycle, and n represents the sequence number of the collected current data, k is the frequency index.

3. The pilot protection method for the outgoing line of a new energy power station according to claim 1, characterized in that, The step of converting the two-dimensional DCT coefficients into binary images specifically includes the following steps: Select the first 64 DCT coefficients and calculate their mean value, and determine the threshold according to the mean value; Compare the DCT coefficients with the threshold. If it is greater than or equal to the threshold, the DCT coefficient is represented as 1. If it is less than the threshold, the DCT coefficient is represented as 0, thereby generating a binary image.

4. A pilot protection method for the outgoing line of a new energy power station according to claim 1, characterized in that, The step of calculating the respective corresponding Hamming distances according to the perceptual hash values, and the Hamming distance is expressed as: ； Among them, h 1 and h 2 are binary hash values, and ⊕ represents a bitwise exclusive OR operation; the Hamming distance is obtained by calculating the number of 1s in the exclusive OR result.

5. A pilot protection system for the outgoing line of a new energy power station, characterized in that, It includes: A collection module for collecting the current data on the new energy power station side and the grid side within one cycle after the occurrence of a fault in the outgoing line of the new energy power station; A calculation module for decomposing the current data on the new energy power station side and the grid side into different frequency components through DCT transformation to obtain the two-dimensional DCT coefficients corresponding to the current data at each time point, and converting the two-dimensional DCT coefficients into binary images; Flatten the binary images of the current data on the new energy power station side and the grid side into one-dimensional vectors, and use them as the respective corresponding perceptual hash values; Calculate the respective corresponding Hamming distances according to each perceptual hash value; A determination module for performing pilot protection by comparing the Hamming distances of the two perceptual hash values; Among them, if the Hamming distance is greater than the set threshold of the in-zone and out-of-zone fault criterion, it is determined that an in-zone fault has occurred in the corresponding phase on the new energy power station side and the grid side, and protection actions are taken; Otherwise, it is determined as an out-of-zone fault and protection restoration is performed.

6. A computer device for the pilot protection of the outgoing line of a new energy power station, characterized in that, It includes: A memory, a processor, and a computer program stored in the memory. When the processor executes the computer program, it implements the steps of the pilot protection method for the outgoing line of the new energy power station according to any one of claims 1-4.

7. A readable storage medium, characterized in that, The readable storage medium stores a computer program, and the computer program includes program instructions. When the program instructions are executed by the processor, they are used to execute the steps of the pilot protection method for the outgoing line of the new energy power station according to any one of claims 1-4.

Images