Directional pilot protection method for centralized configuration flexible direct-current looped network power transmission line
By using the directional longitudinal protection method and the voltage step change feature identification algorithm of the line-mode voltage traveling wave, the problems of reduced fault differentiation and noise interference in centralized flexible DC ring networks are solved, and rapid and reliable identification of faults inside and outside the zone is achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-11-18
- Publication Date
- 2026-04-17
AI Technical Summary
In a centralized flexible DC ring network, traditional protection methods are difficult to effectively distinguish between faults at the end of the protected line and faults at the beginning of the next level line, and are severely affected by noise interference, resulting in inaccurate identification of high-resistance faults.
The directional longitudinal protection method is adopted. By interacting with the fault direction logic signals at both ends, the voltage step change feature identification algorithm of the line-mode voltage traveling wave is used, combined with the backtracking threshold mechanism and the adaptive threshold strategy, to achieve accurate identification of faults inside and outside the zone.
It enables rapid and reliable fault identification of centralized configuration flexible DC ring networks, has strong anti-noise interference capability and transition resistance capability, and can accurately distinguish between faults inside and outside the zone.
Smart Images
Figure CN121886301A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of power system technology and relates to a directional longitudinal protection method for centralized flexible DC ring network transmission lines. Background Technology
[0002] Driven by the "dual carbon" goals, the rapid development of new energy sources has brought new challenges to the safety and stability of the power system. Therefore, the protection methods for flexible DC transmission lines are particularly crucial. They not only relate to the efficient integration of new energy sources and the stable operation of the power system, but also enhance the system's adaptability to fluctuations in new energy supply, ensuring the safety and continuity of power supply.
[0003] However, with the continuous expansion of new energy installed capacity, the traditional method of installing current-limiting reactors at converter station outlets has gradually revealed its drawbacks. On the one hand, current-limiting reactors increase system losses and reduce transmission efficiency; on the other hand, their installation and maintenance in offshore or coastal areas are more difficult, which may affect the stability and dynamic response performance of the system. Therefore, the power grid structure is shifting from a "boundary grid" with distributed current-limiting reactors at line outlets to a new type of power grid with current-limiting reactors centrally configured at converter station outlets.
[0004] In centralized flexible DC ring networks, there are no current-limiting reactors between the protected line and the downstream line, significantly reducing the distinguishability of fault traveling wave signals. Traditional single-ended quantity protection based on boundary elements struggles to effectively differentiate between faults at the end of the protected line and faults at the beginning of the next-level line. Existing single-ended quantity protection schemes for centralized power grids often rely on the distance characteristics of electrical quantities as the criterion. However, this method suffers from dead zones in practical operation, making accurate identification of high-resistance faults inherently difficult, and is severely affected by noise interference. Therefore, a directional longitudinal coordinated protection scheme is adopted to compensate for the shortcomings of single-ended distance criteria, thereby achieving highly reliable fault identification and location. Summary of the Invention
[0005] The purpose of this invention is to provide a directional longitudinal protection method for a centrally configured flexible DC ring network transmission line. This method achieves accurate identification of faults inside and outside the zone by interacting with fault direction logic signals at both ends.
[0006] The technical solution adopted in this invention is a directional longitudinal protection method applicable to centralized flexible DC ring network transmission lines, specifically including the following steps: Step 1, Calculate the line-mode voltage fault component Linear current fault component Then, the traveling wave of the line-mode voltage is calculated based on this. ; Step 2: The voltage step change feature detection algorithm based on the backtracking threshold mechanism identifies the voltage step change feature of the line-mode voltage traveling wave in Step 1. Step 3: Based on the identification results of Step 2, determine the faults inside and outside the area.
[0007] The invention is further characterized by: The specific process of step 1 is as follows: Step 1.1: After the protection starting element is triggered, first mark the start time of the fault identification element as... k = 0, positive voltage is collected at the installation location of the protection device. Negative voltage Positive current and negative current The sampled data were then used to calculate the average positive voltage. Average negative electrode voltage Average positive current and the average value of negative electrode current The data collected at each sampling point after startup are subtracted from the corresponding mean to obtain the positive voltage fault component. Negative voltage fault component Positive current fault component and negative electrode current fault component The calculation process is shown in the following formula (1): (1) In the formula, k Define the time sequence number and the time of fault occurrence. k = 0, 0≤ k ≤ N -1; N This represents the number of sampling points; Step 1.2, obtaining the positive voltage fault component in step 1.1 Negative voltage fault component Positive current fault component and negative current fault component Then, a Kelenberger transform is performed to obtain the line-mode voltage fault components. With line mode current fault component The calculation formula is shown in equation (2): (2) Step 1.3, for line-mode voltage fault components With line mode current fault component Perform traveling wave decomposition to calculate the traveling wave before the line mode voltage. The calculation process is shown in the following formula (3): (3) In the formula, The line-mode impedance; Step 1.4, before obtaining the traveling wave of the line-mode voltage. Then, median filtering is performed on the fault component signal to obtain the filtered data. The calculation formula is as shown in equation (4): (4) In the formula, The filtered first k One sampling point; The original line-mode voltage traveling wave signal; 2 M +1 represents the filter window length; median indicates median operation.
[0008] The specific process of step 2 is as follows: Step 2.1, the line-mode voltage obtained in step 1.2 Extract the maximum amplitude within the data window. And combined with the adjustment coefficient to determine the threshold for voltage step change, the calculation formula is Equation (5): (5) In the formula, This is the adjustment coefficient; Step 2.2: Scan the median-filtered line-mode voltage forward traveling wave within the data window. All sampling points are used to locate the point of minimum amplitude. k min , u fmin ); then from inside the window k min Backtracking point by point until the first satisfied condition is found. – u fmin > u set sampling points k This point is considered to be the initial moment of a voltage step change and is defined as follows: k drop This is the sampled value at the initial moment of the voltage step change, and the detection is terminated.
[0009] In step 2.2, within the data window, the median-filtered line-mode voltage traveling wave sequence is processed. First, determine the point of minimum amplitude. k min , u fmin The calculation formula is shown in equation (6): (6) Then search inside the window k min Corresponding amplitude u fmin = ,by k min Starting from -1, we backtrack point by point, for each candidate sampling point... k Calculate step change value When the first satisfaction is encountered When the sampling point is determined, that point is considered the initial moment of the voltage step change, denoted as . k drop If a voltage step change feature is detected within the window, the backtracking detection ends; if no sampling point that meets the conditions is found after backtracking to the beginning of the sequence, it is determined that there is no voltage step change feature within the window.
[0010] The specific process of step 3 is as follows: If it exists k drop When a voltage step change occurs, it is considered a positive fault and a fault direction logic signal 1 is sent to the other end; otherwise, it does not exist. k drop It identifies the fault as a reverse fault and sends a fault direction logic signal 0 to the other end; it realizes the judgment of faults inside and outside the area based on the fault direction information of the local end and the other end.
[0011] In step 3, if there is a time window k drop If a voltage step change is detected, the fault direction can be determined to be positive, and a fault direction logic signal can be output. X = 1; When no data is detected in the data window k drop or k drop If the fault does not occur within the time window from the fault initiation time to the set time after initiation, the system fault direction can be determined to be in the opposite direction, and a fault direction logic signal can be output. X = 0.
[0012] In step 3, when the fault direction discrimination signal from the other end arrives at the local protection device, the fault is determined to be inside or outside the zone according to the direction logic values of the local and other ends: when the direction logic of both the other end and the local end is 1, it is determined to be an inside-zone fault; when there is a result of non-1 in the direction logic of both ends, it is determined to be an outside-zone fault.
[0013] In step 3, the specific criterion expression is as shown in equation (7): (7) In the formula, X 1 indicates a fault direction logic signal at this end. X 2 represents the logic signal for the fault direction at the other end.
[0014] The beneficial effects of this invention are that it utilizes the characteristic of a significant step change in the first wave of the traveling wave of the fault line mode voltage in a boundless flexible DC ring network to effectively identify the fault direction, thereby constructing a directional longitudinal protection scheme. The method has rapid identification, strong resistance to transition resistance and noise interference, and can distinguish between internal and external faults in the traveling waves of faults at various points in the entire network. Attached Figure Description
[0015] Figure 1 This is a flowchart of the directional longitudinal protection method for a centrally configured flexible vertical ring network according to the present invention; Figure 2 This is a simulation model diagram of a flexible DC power grid with a centralized configuration of current-limiting reactors; Figure 3(a) shows the simulation verification phase, where the transition resistance at Line 1 150 km in the centrally configured flexible DC ring network is 500Ω. The actual line-mode voltage traveling wave waveform measured at the protected measuring point when there is a negative ground fault and is affected by 30 dB Gaussian white noise interference and abnormal flying points. Figure 3(b) shows the simulation verification stage, where the directional longitudinal protection method of this invention for a centralized flexible DC ring network transmission line is applied to a line 150 km in the region with a transition resistance of 500 Ω. A diagram illustrating the voltage step abrupt change characteristics of the line-mode voltage traveling wave under negative grounding fault, and when subjected to 30 dB Gaussian white noise interference and abnormal flying points. Figure 4 During the simulation verification phase, the directional longitudinal protection method of this invention for a centralized flexible DC ring network was applied to achieve a transition resistance of 0 at Line 2 50 km outside the zone (50 km of the line directly below the downstream line outside the zone). Discrimination diagram of voltage step change characteristics of line-mode voltage traveling wave during positive ground fault; Figure 5(a) shows the simulation verification phase, where the transition resistance at Line 4 25 km outside the centralized configuration flexible DC ring network (25 km of the reverse downstream line outside the area) is 500 Ω. The actual line-mode voltage traveling wave waveform measured at the protected measuring point when there is a positive ground fault and is affected by 20dB Gaussian white noise interference and abnormal flying points; Figure 5(b) shows the simulation verification stage, where the directional longitudinal protection method of this invention for centralized flexible DC ring network transmission lines results in a transition resistance of 500 kΩ at Line 4 25 km outside the zone (25 km of the reverse downstream line outside the zone). A diagram illustrating the voltage step abrupt change characteristics of the line-mode voltage traveling wave under positive grounding faults, and when subjected to 20dB Gaussian white noise interference and abnormal flying points. Detailed Implementation
[0016] The following detailed description is provided in conjunction with specific implementation methods.
[0017] Example 1 This invention relates to a directional longitudinal protection method for centrally configured flexible DC ring network transmission lines, specifically comprising the following steps: Step 1: Calculate the line-mode voltage fault component using data from the start-up time to 0.2 ms after start-up. Linear current fault component Then, the traveling wave of the line-mode voltage is calculated based on this. ; Step 2: The voltage step change feature detection algorithm based on the backtracking threshold mechanism identifies the voltage step change feature of the line-mode voltage traveling wave in Step 1. Step 3, if it exists k drop When a voltage step change occurs, it is considered a positive fault and a fault direction logic signal 1 is sent to the other end; otherwise, it does not exist. k drop It identifies the fault as a reverse fault and sends a fault direction logic signal 0 to the other end; it realizes the judgment of faults inside and outside the area based on the fault direction information of the local end and the other end.
[0018] Example 2 The specific process of step 1 is as follows: Step 1.1: After the protection starting element is triggered, first mark the start time of the fault identification element as... k = 0, the positive voltage is collected at the installation location of the protection device from the start time to 0.2 ms after start-up. Negative voltage Positive current and negative current The sampling data was then used. Subsequently, the average positive voltage during the 10ms period prior to protection activation was calculated. Average negative electrode voltage Average positive current and the average value of negative electrode current The data collected at each sampling point after startup are subtracted from the corresponding mean to obtain the positive voltage fault component. Negative voltage fault component Positive current fault component and negative electrode current fault component This processing method can effectively eliminate the influence of the steady-state fundamental component and extract the transient component that reflects the fault characteristics. The calculation relationship is given by formula (1): (1) In the formula, k Define the time sequence number and the time of fault occurrence. k = 0, 0≤ k ≤ N -1; N This represents the number of sampling points from the moment the faulty component starts up to 0.2 ms after startup.
[0019] Step 1.2, obtaining the positive voltage fault component in step 1.1 Negative voltage fault component Positive current fault component and negative current fault component Next, a Kelvin-Raphson transform is performed on the above data. This transform converts the positive and negative extreme values into linear moduli, thereby obtaining the linear moduli voltage fault components. With line mode current fault component This calculation process can effectively eliminate the influence of interpolar coupling, and its calculation formula is shown in equation (2): (2) Step 1.3: After obtaining the line-mode voltage fault component, the data is further processed by traveling wave decomposition to calculate the line-mode voltage forward traveling wave. The specific calculation relationship is shown in formula (3).
[0020] (3) In the formula, The impedance of the line mode wave is given.
[0021] Step 1.4, before obtaining the traveling wave of the line-mode voltage. Subsequently, to suppress random noise and flying-point interference and improve the accuracy of transient component feature extraction, median filtering was performed on the aforementioned fault component signals to obtain filtered data. Its calculation formula is as shown in equation (4): (4) In the formula, The filtered first k One sampling point; The original line-mode voltage traveling wave signal; 2 M +1 represents the filter window length, M is 2; median indicates median operation.
[0022] Example 3 The specific process of step 2 is as follows: Step 2.1: To further prevent transition resistance and noise interference, an adaptive threshold strategy is adopted when querying whether a voltage step change has occurred: The line-mode voltage obtained in Step 1.2... Extract the maximum amplitude within the data window. And combined with the adjustment coefficient to determine the threshold for voltage step change, the calculation formula is Equation (5): (5) In the formula, This is the adjustment coefficient. Considering the positive correlation and certain proportion between the voltage step change amplitude and the maximum amplitude of the line-mode voltage, and to ensure accurate identification of voltage step change characteristics even under noise interference, this paper takes... =0.25.
[0023] Step 2.2: Scan the median-filtered line-mode voltage forward traveling wave within the data window. All sampling points are used to locate the point of minimum amplitude. k min , u fmin ); then from inside the window k min Backtracking point by point until the first satisfied condition is found. – u fmin > u set sampling points k This point is considered the initial point of the voltage step change and is defined as follows: k drop This is the sampled value at the initial point of the voltage step change, and the detection is terminated.
[0024] Within the data window, the median-filtered line-mode voltage traveling wave sequence is... First, determine its minimum point. k min The calculation formula is shown in equation (6): (6) And find k min Corresponding amplitude u fmin = ,by k min Starting from -1, we backtrack point by point, for each candidate sampling point... k Calculate mutation value When the first satisfaction is encountered When the sampling point is determined, that point is considered the initial moment of the voltage step change, denoted as . k drop = k If the sampling point that meets the conditions is not found when tracing back to the beginning of the sequence, it is determined that "no voltage step change feature occurs within the window" and the backtracking detection ends; if no sampling point that meets the conditions is found when tracing back to the beginning of the sequence, it is determined that "no voltage step change feature occurs within the window".
[0025] Example 4 The specific process of step 3 is as follows: Step 3.1, if there is a time window k drop If a voltage step change is detected, the fault direction can be determined to be positive, and a fault direction logic signal can be output. X = 1; When no data is detected in the data window k drop or k drop If the fault does not occur within the time window from the fault initiation time to 0.2 ms after initiation, the fault direction can be determined to be reversed, and a fault direction logic signal can be output. X = 0; Step 3.2: When the fault direction discrimination signal from the other end reaches the local protection device, the fault is determined to be either inside or outside the protection zone based on the direction logic values of the local and other ends: when the direction logic values of both ends are 1 (both determine that the fault is located within the protected section), it is determined to be an inside-zone fault; when there is a non-1 result in the direction logic values of both ends (one end determines that the fault is located in the opposite direction), it is determined to be an outside-zone fault. The specific criterion expression is as shown in equation (7): (7) In the formula, X 1 indicates a fault direction logic signal at this end. X 2 represents the logic signal for the fault direction at the other end.
[0026] As attached Figure 2 The figure shown is a simulation model of a centralized flexible DC power grid. The figure includes a current-limiting reactor. L 1~ L All four parameters are 0.2 H. MMC1, MMC2, MMC3, and MMC4 are Modular Multilevel Converters (MMCs). P1 is the line boundary protection measuring point. The positive direction of the protection measuring point is defined as from the busbar to the protected line. (See diagram.) f MMC1 , f MMC2 , f MMC3 , f MMC4 These represent faults occurring at the output of the corresponding converter station. f Line11 , f Line12 , f Line13 These represent faults occurring at the beginning, middle, and end of Line 1, respectively. fLine21 , f Line22 , f Line23 These represent faults occurring at the beginning, middle, and end of Line 2, respectively. f Line31 , f Line32 , f Line33 These represent faults occurring at the beginning, middle, and end of Line 3, respectively. f Line41 , f Line42 , f Line43 These represent faults occurring at the beginning, section, and end of Line 4, respectively. Note that at the line boundaries, the point closer to the protection measuring point P1 is considered the beginning, and the point farther from the protection measuring point is considered the end, regardless of the actual distance between the line boundaries and the protection measuring points. This system has a rated voltage of ±500 kV, a rated transmission capacity of 3000 MVA, and transmission lines Line 1 is 207.9 km long, Line 2 is 192.7 km long, Line 3 is 217.2 km long, and Line 4 is 49.9 km long. An overhead line frequency-varying parameter model is used, and its line surge impedances are Z... c0 =320 Z c1 =260 During system simulation, the sampling frequency was 100 kHz, and a 1.2 / 50 μs standard lightning current model was used to simulate lightning strikes. The entire text uses Line 1, measuring point P1 as an example; the fault distance is the distance from the fault location to measuring point P1, as shown in the attached figure. Figure 1 The identification process.
[0027] Example 5 As shown in Figures 3(a) and 3(b), in the case of Figure 2 In the centralized flexible DC grid configuration shown, a transition resistance of 500 kΩ occurs at Line 1150 km within the region. Simulated values of the line-mode voltage traveling wave measured at the protection measuring point under negative ground fault conditions, and affected by 30 dB Gaussian white noise interference and abnormal flying points. After the protection element starts, the line-mode voltage traveling wave data of 20 sampling points from the start time to after start-up are read. Calculation. k drop Based on the fault direction criterion, the fault direction is determined to be positive. Combined with the signal 1 from the opposite direction element, it is determined to be a fault within the zone.
[0028] Example 6 like Figure 4 As shown, in Figure 2In the centralized flexible DC grid configuration shown, a transition resistance of 0 occurs at Line 2 50 km outside the region (50 km of the direct downstream line from outside the region). Simulated value of the line-mode voltage traveling wave measured at the protection measuring point during a positive ground fault. After the protection element starts, read the line-mode voltage traveling wave data from the start time to 20 sampling points after start-up. Calculate. k drop Based on the fault direction criterion, the fault direction is determined to be positive. Combined with the signal 0 from the opposite direction element, it is determined to be an external fault.
[0029] Example 7 As shown in Figures 5(a) and 5(b), in the case of Figure 2 In the centralized flexible DC grid configuration shown, a transition resistance of 500 kΩ occurs at Line 425 km outside the region (25 km of the reverse downstream line outside the region). The simulated value of the line-mode voltage traveling wave measured at the protection measuring point under a positive ground fault, and affected by 20dB Gaussian white noise interference and abnormal flying points. After the protection element starts, the line-mode voltage traveling wave data of 20 sampling points from the start time to after start-up are read. Calculation. k drop Based on the fault direction criterion, the fault direction is determined to be in the opposite direction, indicating that it is an external fault.
[0030] To comprehensively verify the impact of fault distance, lightning strike, transition resistance, and system type on the judgment results, a four-terminal centralized configuration flexible DC ring network was used as an example, with lightning strike faults and transition resistance set to 0. 500 The grounding fault was identified, and the identification method was verified based on the simulation results.
[0031] Table 1 provides simulation data for a four-terminal centralized flexible DC ring network at different fault distances, where all faults are metallic grounding faults. The data shows that faults inside and outside the fault zone can be distinguished at different fault distances.
[0032] Table 1 Simulation verification results for different fault distances
[0033] Table 2 shows the simulation data for different types of faults in the four-terminal centralized configuration flexible DC ring network. The data shows that faults inside and outside the zone can be distinguished under different fault types.
[0034] Table 2 Simulation verification results under different fault types
[0035] This invention normalizes the traveling wave of the line-mode voltage and identifies the fault direction by extracting the voltage step change characteristics of the traveling wave. This method also boasts advantages such as low computational complexity and fast response speed, meeting the real-time and high-precision requirements of flexible DC transmission systems. The purpose of this invention is to provide a directional longitudinal protection scheme for centrally configured flexible DC ring network transmission lines. This method utilizes the significant differences in the time-domain waveforms of the traveling wave during forward and reverse faults, employing an adaptive threshold to detect the voltage step change characteristics of the reverse traveling wave of the line-mode voltage. This reduces the impact of transition resistance on the identification method, exhibits strong anti-interference capabilities, and overcomes the maloperational defects of existing protection methods based on boundary conditions in centrally configured flexible DC ring networks.
[0036] This invention discloses a directional longitudinal protection method for centrally configured flexible DC ring network transmission lines. After the protection element is activated, the method extracts transient electrical components and obtains the line-mode voltage and current. The method calculates the forward traveling wave of the line-mode voltage through traveling wave decomposition and uses median filtering to suppress Gaussian white noise and flying point interference, thereby obtaining stable and reliable transient information. Subsequently, a voltage step change criterion based on an adaptive threshold is used to identify voltage step change characteristics. A backtracking threshold algorithm is used to determine the initial moment of the voltage step change, and a time window condition is combined to determine the fault direction. After the directional signals from the local and remote ends interact, faults within and outside the fault zone can be accurately distinguished, achieving fast and reliable protection for centrally configured flexible DC ring network transmission lines. This invention has advantages such as low computational load, fast response speed, strong resistance to transition resistance, and strong noise immunity. It can effectively solve the misjudgment problem existing in traditional protection methods relying on boundary elements in centrally configured flexible DC ring network transmission lines, providing a guarantee for the safe and stable operation of flexible DC ring networks.
Claims
1. A directional longitudinal protection method for a centrally configured flexible vertical ring network, characterized in that: Specifically, the steps include the following: Step 1, Calculate the line-mode voltage fault component Linear current fault component Then, the traveling wave of the line-mode voltage is calculated based on this. ; Step 2: The voltage step change feature detection algorithm based on the backtracking threshold mechanism identifies the voltage step change feature of the line-mode voltage traveling wave in Step 1. Step 3: Based on the identification results of Step 2, determine the faults inside and outside the area.
2. The directional longitudinal protection method for a centrally configured flexible vertical ring network according to claim 1, characterized in that: The specific process of step 1 is as follows: Step 1.1: After the protection starting element is triggered, first mark the start time of the fault identification element as... k = 0, positive voltage is collected at the installation location of the protection device. Negative voltage Positive current and negative current The sampled data were then used to calculate the average positive voltage. Average negative electrode voltage Average positive current and the average value of negative electrode current The data collected at each sampling point after startup are subtracted from the corresponding mean to obtain the positive voltage fault component. Negative voltage fault component Positive current fault component and negative electrode current fault component The calculation process is shown in the following formula (1): （1） In the formula, k Define the time sequence number and the time of fault occurrence. k = 0, 0≤ k ≤ N -1; N This represents the number of sampling points; Step 1.2, obtaining the positive voltage fault component in step 1.1 Negative voltage fault component Positive current fault component and negative current fault component Then, a Kelenberger transform is performed to obtain the line-mode voltage fault components. With line mode current fault component The calculation formula is shown in equation (2): （2） Step 1.3, for line-mode voltage fault components With line mode current fault component Perform traveling wave decomposition to calculate the traveling wave before the line mode voltage. The calculation process is shown in the following formula (3): （3） In the formula, The line-mode impedance; Step 1.4, before obtaining the traveling wave of the line-mode voltage. Then, median filtering is performed on the fault component signal to obtain the filtered data. The calculation formula is as shown in equation (4): （4） In the formula, The filtered first k One sampling point; The original line-mode voltage traveling wave signal; 2 M +1 represents the filter window length; median indicates median operation.
3. The directional longitudinal protection for a centrally configured flexible vertical ring network according to claim 2, characterized in that: The specific process of step 2 is as follows: Step 2.1, the line-mode voltage obtained in step 1.2 Extract the maximum amplitude within the data window. And combined with the adjustment coefficient to determine the threshold for voltage step change, the calculation formula is Equation (5): （5） In the formula, This is the adjustment coefficient; Step 2.2: Scan the median-filtered line-mode voltage forward traveling wave within the data window. All sampling points are used to locate the point of minimum amplitude. k min , u fmin ); then from inside the window k min Backtracking point by point until the first satisfied condition is found. – u fmin > u set sampling points k This point is considered to be the initial moment of a voltage step change and is defined as follows: k drop This is the sampled value at the initial moment of the voltage step change, and the detection is terminated.
4. The directional longitudinal protection method for a centrally configured flexible vertical ring network according to claim 3, characterized in that: In step 2.2, within the data window, the median-filtered line-mode voltage traveling wave sequence is processed. First, determine the minimum point. k min The calculation formula is shown in equation (6): （6） Find k min Corresponding amplitude u fmin = ,by k min Starting from -1, we backtrack point by point, for each candidate sampling point... k Calculate drop value When the first satisfaction is encountered When a sampling point is selected, that point is considered the starting point of the voltage step change, denoted as . k drop = k If a voltage step change feature is detected within the window, the backtracking detection ends; if no sampling point that meets the conditions is found after backtracking to the beginning of the sequence, it is determined that there is no voltage step change feature within the window.
5. The directional longitudinal protection method for a centrally configured flexible vertical ring network according to claim 4, characterized in that: The specific process of step 3 is as follows: If there exists k drop When a voltage step change occurs, it is considered a positive fault and a fault direction logic signal 1 is sent to the other end; otherwise, it does not exist. k drop It identifies the fault as a reverse fault and sends a fault direction logic signal 0 to the other end; it realizes the judgment of faults inside and outside the area based on the fault direction information of the local end and the other end.
6. The directional longitudinal protection method for a centrally configured flexible vertical ring network according to claim 5, characterized in that: In step 3, if there is a time window k drop If a voltage step change occurs, the fault direction can be determined to be positive, and a fault direction logic signal can be output. X = 1; When no data is detected in the data window k drop or k drop If the fault does not occur within the time window from the fault initiation time to the set time after initiation, the system fault direction can be determined to be in the opposite direction, and a fault direction logic signal can be output. X = 0.
7. The directional longitudinal protection method for a centrally configured flexible vertical ring network according to claim 6, characterized in that: In step 3, when the fault direction discrimination signal from the other end arrives at the local protection device, the fault is determined to be inside or outside the zone according to the direction logic values of the local and other ends: when the direction logic of both the other end and the local end is 1, it is determined to be an inside-zone fault; when there is a result of non-1 in the direction logic of both ends, it is determined to be an outside-zone fault.
8. The directional longitudinal protection method for a centrally configured flexible vertical ring network according to claim 7, characterized in that: In step 3, the specific criterion expression is as shown in equation (7): （7） In the formula, X 1 indicates a fault direction logic signal at this end. X 2 represents the logic signal for the fault direction at the other end.