Power distribution network fault location method and system based on CSA-DDT optimization

Abstract

The application relates to the technical field of power system distribution network fault detection and positioning, and particularly discloses a distribution network fault distance measurement method and system based on CSA-DDT optimization. The application extracts zero-mode and line-mode current anti-travel wave signals by collecting three-phase transient currents and three-phase transient voltages of each monitoring point of the distribution network and performing phase-mode transformation, globally optimizes the continuous smooth window width, threshold coefficient and continuous over-threshold time of a differential dynamic transient operator by using a crow search algorithm to obtain optimal parameters, substitutes the optimal parameters into the differential dynamic transient operator to extract zero-mode and line-mode wave arrival sampling points of each monitoring point and calculate a time difference to determine a reference node, constructs a double-end distance measurement equation to solve a fault distance estimation value in combination with a zero-mode wave speed and a nonlinear relationship formula of the fault distance, and performs weighted averaging, so that the precision and anti-interference capability of complex distribution network fault distance measurement are improved.

Description

Technical Field

[0001] This invention relates to the field of power system distribution network fault detection and location technology, and in particular to a distribution network fault location method and system based on CSA-DDT optimization. Background Technology

[0002] High-precision and high-reliability fault location methods for distribution networks are crucial for improving the operation and maintenance level of distribution networks. Traditional fault location typically employs the impedance method, calculating the faulty line impedance using line parameters and measured values ​​of power frequency voltage and current at fault nodes. However, with increasingly complex power grid structures, distribution lines generally exhibit multi-branch and multi-terminal characteristics. When a fault occurs, current is shunted among the branches, causing changes in the system's equivalent impedance and current distribution. This makes it difficult for single-ended voltage and current measurements to accurately reflect the impedance characteristics of the fault point. Furthermore, the impedance method is highly susceptible to the effects of transition resistance, line load fluctuations, and changes in system operating modes. In scenarios involving high-resistance grounding faults or short-line faults, there is a location dead zone, leading to significant location errors or even ineffective fault location.

[0003] In contrast, the traveling wave method uses high-frequency transient traveling wave information generated at the moment of fault occurrence for location, and is less affected by transition resistance and system operation mode. However, in the multi-branch structure of the distribution network, the traditional single-ended traveling wave method is prone to the superposition of reflected waves and initial wavefronts, making reflected wave identification difficult and susceptible to wave velocity uncertainty and waveform distortion. While the double-ended traveling wave method improves accuracy to some extent, it is difficult to reliably determine the branch where the fault is located in complex structures.

[0004] Therefore, multi-terminal traveling wave fault location methods have gradually become an important direction for fault location in distribution networks because they can comprehensively utilize the arrival time information of traveling waves from multiple measurement points, effectively reduce interference from traveling wave reflection and wavefront aliasing, and combine topology information to determine fault branches. Even with the introduction of empirical mode decomposition, variational mode decomposition algorithms, or deep learning networks for wavefront identification in recent years, it is still difficult to balance the accuracy, robustness, and engineering practicality of wavefront extraction when facing complex field conditions such as multi-branch topology, wave velocity dispersion effects, and strong noise interference.

[0005] In summary, in the field of multi-terminal traveling wave ranging in complex distribution networks, there is currently a lack of an effective and highly accurate wavefront detection and arrival time extraction method with strong anti-interference capabilities. Furthermore, existing wavefront identification and protection measures cannot completely solve the problems of limited ranging accuracy caused by strong noise, wave velocity dispersion, and excessive reliance on subjective parameters and clock synchronization in multi-branch topologies. Summary of the Invention

[0006] To address the aforementioned technical problems, this invention provides a method and system for fault location in distribution networks based on CSA-DDT optimization.

[0007] In a first aspect, the present invention provides a distribution network fault location method based on CSA-DDT optimization, the technical solution of which is as follows: The three-phase transient current and three-phase transient voltage at each monitoring point of the distribution network are collected. Based on the three-phase transient current and three-phase transient voltage, phase mode transformation is performed and reverse traveling waves are extracted to obtain zero-mode current reverse traveling wave signals and line-mode current reverse traveling wave signals. The crow search algorithm is used to globally optimize the continuous smooth window width, threshold coefficient, and duration of overthreshold for the differential dynamic transient operator. In each iteration, the fitness value is calculated based on a set of parameters consisting of the continuous smooth window width, the threshold coefficient, and the duration of overthreshold. The crow position is updated with the goal of minimizing the fitness value, so as to obtain the optimal continuous smooth window width, the optimal threshold coefficient, and the optimal duration of overthreshold. Substituting the optimal continuous smooth window width, the optimal threshold coefficient, and the optimal sustained overthreshold time into the differential dynamic transient operator, the arrival sampling points of the zero-mode current reverse traveling wave signal and the line-mode current reverse traveling wave signal are extracted respectively. Based on the zero-mode arrival sampling points and line-mode arrival sampling points extracted for each monitoring point, the time difference between the zero-mode arrival time and the line-mode arrival time of the monitoring point is calculated. A time difference vector is constructed based on the time differences of all monitoring points, and the monitoring point corresponding to the smallest element in the time difference vector is determined as the reference node. Obtain the pre-fitted nonlinear relationship between zero-mode wave velocity and fault distance, construct a two-end ranging equation with the reference node and each auxiliary node respectively, and solve the fault distance estimate corresponding to each auxiliary node according to the nonlinear relationship and the two-end ranging equation. The estimated fault distance is weighted and averaged to obtain the final fault distance.

[0008] The beneficial effects of the distribution network fault location method based on CSA-DDT optimization of the present invention are as follows: The method of this invention collects three-phase transient currents and three-phase transient voltages at various monitoring points in the distribution network and performs phase-mode transformation to extract zero-mode and line-mode current reverse traveling wave signals. It uses a crow search algorithm to globally optimize the continuous smoothing window width, threshold coefficient, and duration of over-threshold time of the differential dynamic transient operator to obtain the optimal parameter combination. The optimal parameters are then substituted into the differential dynamic transient operator to extract the zero-mode and line-mode wave arrival sampling points at each monitoring point and calculate the time difference to determine the reference node. A two-end ranging equation is constructed by combining a pre-fitted nonlinear relationship between zero-mode wave velocity and fault distance to solve for the estimated fault distance, and a weighted average is performed. This method solves the problems of excessive dead zones and ranging errors caused by transition resistance, load fluctuations, and changes in operating modes in multi-branch dominated power grids, as well as the insufficient accuracy and robustness of wavefront extraction caused by reflected wave aliasing, wavefront identification difficulties, wave velocity uncertainty, and strong noise interference in multi-branch topologies using the traditional impedance method. This improves the accuracy and anti-interference capability of fault ranging in complex distribution networks.

[0009] Based on the above scheme, the distribution network fault location method based on CSA-DDT optimization of the present invention can be further improved as follows.

[0010] In one alternative approach, the steps of performing phase-mode transformation based on the three-phase transient currents and the three-phase transient voltages and extracting the reverse traveling wave to obtain the zero-mode current reverse traveling wave signal and the line-mode current reverse traveling wave signal include: The three-phase transient current and the three-phase transient voltage are subjected to phase-mode transformation using the Karrenbauer transformation matrix, which is: ;in, This represents the phase-mode transformation matrix, used to transform the three-phase transient current and three-phase transient voltage into zero-mode components, first line-mode components, and second line-mode components, respectively. The Begeron algorithm is used to extract the inverse traveling wave from the zero-mode component, the first line-mode component, and the second line-mode component, respectively; wherein, the Begeron algorithm is based on the fault voltage mixed wave signal. Fault current mixed wave signal and line impedance Calculate the reverse traveling wave of the current The calculation formula is: ; The zero-mode current reverse traveling wave signal is obtained based on the zero-mode component, the first line-mode current reverse traveling wave signal is obtained based on the first line-mode component, and the second line-mode current reverse traveling wave signal is obtained based on the second line-mode component; the first line-mode current reverse traveling wave signal and the second line-mode current reverse traveling wave signal are averaged to obtain an average line-mode current reverse traveling wave signal, which is used as the line-mode current reverse traveling wave signal.

[0011] The advantages of adopting the above-mentioned optional method are as follows: further performing phase-mode transformation on the three-phase transient current and voltage through the Karrenbauer transformation matrix to separate the zero-mode component and two line-mode components, using the Begeron algorithm to extract the reverse traveling wave of each mode current respectively, and averaging the two line-mode current reverse traveling waves to obtain the average line-mode current reverse traveling wave signal, thereby improving the accuracy of mode separation and the stability of reverse traveling wave extraction, and providing a balanced line-mode signal basis for subsequent wave arrival time detection.

[0012] In one alternative approach, the steps for calculating the fitness value include: The jog energy is calculated based on the zero-mode current reverse traveling wave signal and the line-mode current reverse traveling wave signal. Regarding the jump energy Normalization is performed to obtain normalized energy. The normalized energy is obtained by using a Gaussian smoothing kernel. Convolution smoothing yields a smoothed energy curve. According to the threshold coefficient With the smooth energy curve peak Determine the detection threshold This will be the first time that the threshold for sustained proportion is met. The sampling point corresponding to the time is used as the arrival sampling point. The duration of exceeding the threshold is: ;in, This is an indicator function that takes the value 1 when the condition is true and 0 otherwise. fitness value The calculation formula is: .

[0013] The advantages of adopting the above optional method are as follows: the juxtaposition energy is further calculated based on the reverse traveling wave of the zero-mode and line-mode currents and normalized. The normalized energy is convolved and smoothed using a Gaussian smoothing kernel to obtain a smooth energy curve. The detection threshold is determined based on the threshold coefficient and the peak value. The arrival sampling point is determined by combining the over-threshold duration ratio. The detection quality is measured by the fitness function, which improves the wavefront identification accuracy and the reliability of wave arrival time extraction.

[0014] In one alternative approach, the Gaussian smoothing kernel is constructed as follows: based on the sampling frequency Continuous smooth window width Converting from time scale to sampling point scale ,Will The standard deviation of the Gaussian smoothing kernel is calculated using the full width at half maximum (FWHM). In discrete networks Construct a normalized Gaussian smoothing kernel: in, To smooth the kernel half-width, the value is [value to be filled in]. .

[0015] The advantages of adopting the above optional method are as follows: the continuous smoothing window width is further converted from the time scale to the sampling point scale according to the sampling frequency, the standard deviation of the Gaussian smoothing kernel is calculated with the full width at half height, a normalized Gaussian smoothing kernel is constructed on the discrete network and the half width of the smoothing kernel is determined, which ensures the adaptability of the smoothing kernel parameters to the sampling frequency and improves the continuity and boundary accuracy of the energy curve smoothing process.

[0016] In one alternative approach, the number of continuously exceeding the threshold sampling points According to the duration of the overthreshold and sampling frequency Confirmed, the calculation formula is as follows: .

[0017] The advantages of adopting the above optional method are as follows: further determining the number of continuous over-threshold sampling points based on the continuous over-threshold time and sampling frequency, directly converting the time parameter into the number of discrete sampling points, making the over-threshold duration ratio determination strictly correspond to the digital sampling sequence, enhancing the compatibility between continuous over-threshold determination and the sampling process, and improving the timeliness consistency of the arrival sampling point detection.

[0018] In one alternative approach, when performing global optimization using a crow search algorithm, the crow population size is assumed to be... The maximum number of iterations is The probability of perception is The flight step length coefficient is ; In each iteration, for the th One crow, randomly select another different crow Generate random numbers that follow a uniform distribution. ;like Then the first The location of the crow has been updated to ,in For crows The historical best memory location; if Then the first The location of the crow has been updated to ,in and These are the lower and upper bounds of the search space, respectively.

[0019] The beneficial effects of adopting the above optional method are as follows: by further setting the population size, maximum number of iterations, perception probability and flight step size coefficient, and selecting to update the crow position by following the historical best memory position or by random search based on the relationship between the random number and the perception probability, the global optimization is carried out with the goal of minimizing the fitness value, which improves the convergence speed and global search capability of the parameter optimization process.

[0020] In one alternative approach, the nonlinear relationship between the zero-mode wave velocity and the fault distance is a fractional function: in, For monitoring points Zero-mode wave velocity at that location From the fault point to the monitoring point distance, , , , These are the constant parameters obtained through pre-fitting.

[0021] The advantages of adopting the above-mentioned optional method are as follows: further using a fractional function to fit the nonlinear relationship between zero-mode wave velocity and fault distance, using pre-calibrated constant parameters to describe the variation law of zero-mode wave velocity under different fault distances, substituting the nonlinear relationship into the two-end ranging equation, improving the accuracy of zero-mode wave velocity value, and providing a reliable wave velocity basis for fault distance calculation.

[0022] In one alternative approach, when performing a weighted average of the fault distance estimates, the fault distance estimate for each auxiliary node is... weight Reference node To the auxiliary node distance The reciprocal of, that is ; The weights are normalized to obtain normalized weights. ,in The total number of auxiliary nodes; the final fault distance is: .

[0023] The beneficial effects of adopting the above optional method are as follows: by further using the reciprocal of the distance from the reference node to the auxiliary node as the weight of the fault distance estimate of each auxiliary node, and then normalizing the weights and averaging them, the interference of the estimation error of the long-distance auxiliary node on the final result is reduced, and the accuracy of the comprehensive fault distance estimation and the rationality of the contribution of each node are improved.

[0024] In one alternative approach, the two-end ranging equation constructed by the reference node and each auxiliary node is: in, Reference node The time difference between the zero-mode wave arrival time and the line-mode wave arrival time. as auxiliary node The time difference mentioned above, The reference node From the fault point to the auxiliary node The shortest distance, From the fault point to the reference node distance, For linear mode wave velocity, and Determined based on the aforementioned nonlinear relationship.

[0025] The advantages of adopting the above-mentioned optional approach are as follows: it further constructs a multivariate two-end ranging equation that includes the time difference ratio, fault distance ratio, and the relationship between the line mode and the zero mode wave velocity, makes full use of the difference in the zero mode wave arrival time and the nonlinear wave velocity relationship between the reference node and the auxiliary node, improves the rigor of multi-end information fusion ranging, and enhances the accuracy of fault distance calculation under complex topology.

[0026] Secondly, this invention provides a distribution network fault location system optimized based on CSA-DDT, the technical solution of which is as follows: The processing module is used to collect three-phase transient current and three-phase transient voltage at each monitoring point of the distribution network, perform phase mode transformation based on the three-phase transient current and three-phase transient voltage, and extract the reverse traveling wave to obtain zero-mode current reverse traveling wave signal and line-mode current reverse traveling wave signal; The iterative module is used to globally optimize the continuous smooth window width, threshold coefficient, and duration of overthreshold time of the differential dynamic transient operator using the crow search algorithm. In each iteration, the fitness value is calculated based on a set of parameters consisting of the continuous smooth window width, the threshold coefficient, and the duration of overthreshold time. The crow position is updated with the goal of minimizing the fitness value, so as to obtain the optimal continuous smooth window width, the optimal threshold coefficient, and the optimal duration of overthreshold time. The determination module is used to substitute the optimal continuous smooth window width, the optimal threshold coefficient, and the optimal sustained overthreshold time into the differential dynamic transient operator, extract the arrival sampling points of the zero-mode current reverse traveling wave signal and the line-mode current reverse traveling wave signal respectively, calculate the time difference between the zero-mode arrival time and the line-mode arrival time of the monitoring point based on the zero-mode arrival sampling points and the line-mode arrival sampling points extracted for each monitoring point, construct a time difference vector based on the time differences of all monitoring points, and determine the monitoring point corresponding to the smallest element in the time difference vector as the reference node; The calculation module is used to obtain the pre-fitted nonlinear relationship between zero-mode wave velocity and fault distance, construct a two-end ranging equation with the reference node and each auxiliary node respectively, and solve the fault distance estimate corresponding to each auxiliary node according to the nonlinear relationship and the two-end ranging equation. The output module is used to perform weighted averaging on the estimated fault distance to obtain the final fault distance.

[0027] The beneficial effects of the distribution network fault location system based on CSA-DDT optimization of the present invention are as follows: The system of this invention collects three-phase transient currents and three-phase transient voltages at various monitoring points in the distribution network and performs phase-mode transformation to extract zero-mode and line-mode current reverse traveling wave signals. It uses a crow search algorithm to globally optimize the continuous smoothing window width, threshold coefficient, and duration of over-threshold time of the differential dynamic transient operator to obtain the optimal parameter combination. The optimal parameters are then substituted into the differential dynamic transient operator to extract the zero-mode and line-mode wave arrival sampling points at each monitoring point and calculate the time difference to determine the reference node. A two-end ranging equation is constructed by combining a pre-fitted nonlinear relationship between zero-mode wave velocity and fault distance to solve for the estimated fault distance, and a weighted average is performed. This solves the problems of excessive dead zones and ranging errors caused by transition resistance, load fluctuations, and changes in operating modes in multi-branch dominated power grids, as well as the insufficient accuracy and robustness of wavefront extraction caused by reflected wave aliasing, wavefront identification difficulties, wave velocity uncertainty, and strong noise interference in multi-branch topologies of the traditional impedance method. This improves the accuracy and anti-interference capability of fault ranging in complex distribution networks.

[0028] Thirdly, the technical solution of an electronic device according to the present invention is as follows: It includes a memory, a processor, and a program stored in the memory and running on the processor, wherein the processor executes the program to implement the steps of the distribution network fault location method based on CSA-DDT optimization of the present invention.

[0029] Fourthly, the technical solution of a computer-readable storage medium provided by the present invention is as follows: The computer-readable storage medium stores instructions that, when read, cause the computer-readable storage medium to perform the steps of the distribution network fault location method based on CSA-DDT optimization of the present invention.

[0030] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, and in order to make the above and other objects, features and advantages of the present invention more apparent and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0031] The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings: Figure 1 This is a flowchart illustrating an embodiment of a distribution network fault location method based on CSA-DDT optimization according to the present invention. Figure 2 This is a schematic diagram of the network topology of a 35kV fault distribution network. Figure 3 This is a schematic diagram comparing modulus-inverse traveling waves and mixed waves; Figure 4 This is a schematic diagram of the convergence curve of the crow search algorithm; Figure 5 This is a schematic diagram of the process of extracting sampling points in Porta. Figure 6 A schematic diagram of the fitting curve of the zero-mode wave velocity as a function of fault distance; Figure 7 This is a schematic diagram of an embodiment of a distribution network fault location system based on CSA-DDT optimization according to the present invention; Figure 8 This is a schematic diagram of an embodiment of an electronic device according to the present invention. Detailed Implementation

[0032] Exemplary embodiments of the invention will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention can be implemented in various forms and should not be limited to the embodiments set forth herein.

[0033] Figure 1 This diagram illustrates a flowchart of an embodiment of a CSA-DDT-optimized distribution network fault location method provided by the present invention. This CSA-DDT-optimized distribution network fault location method can be executed by electronic devices such as terminal devices or servers. The terminal device can be any fixed or mobile terminal, such as user equipment (UE), mobile device, user terminal, terminal, cellular phone, cordless phone, personal digital assistant (PDA), handheld device, computing device, vehicle-mounted device, or wearable device. The server can be a single server or a server cluster consisting of multiple servers. Any electronic device can implement the CSA-DDT-optimized distribution network fault location method by having its processor call computer-readable instructions stored in its memory. Figure 1 As shown, it includes the following steps: S1. Collect the three-phase transient current and three-phase transient voltage at each monitoring point of the distribution network, perform phase mode transformation based on the three-phase transient current and three-phase transient voltage, and extract the reverse traveling wave to obtain the zero-mode current reverse traveling wave signal and the line-mode current reverse traveling wave signal.

[0034] In this context, the distribution network refers to a power distribution network composed of distribution substations, feeders, branch lines, and user access equipment, used to reduce high-voltage power to medium and low voltage and distribute it to end users. For example, a 35kV radial distribution network includes a central bus N6 and five line terminals N1, N2, N3, N4, and N5, with the fault point F located on the line between N5 and N6. A monitoring point refers to the physical location in the distribution network where a traveling wave monitoring device is installed to collect transient fault signals. For example, traveling wave monitoring devices are installed at nodes N1, N2, N3, N4, and N5, with each device simultaneously collecting three-phase transient current and three-phase transient voltage.

[0035] Among them, the three-phase transient current refers to the instantaneous changing current signals generated by phases A, B, and C in a power line when a fault occurs; for example, the three-phase transient currents collected at node N5 are respectively , , Three-phase transient voltage refers to the instantaneous voltage changes in phases A, B, and C of a power line when a fault occurs; for example, the three-phase transient voltages collected at node N5 are respectively... , , .

[0036] Phase-mode transformation refers to the mathematical transformation that decouples three-phase coupled electrical quantities into independent modulus components using a transformation matrix; for example, using the Karrenbauer transformation matrix. The three-phase transient current and three-phase transient voltage are transformed into zero-mode components, first-line-mode components, and second-line-mode components, respectively. A reverse traveling wave refers to a wave component in a traveling wave that propagates from the fault point to the line monitoring point in the opposite direction to the forward traveling wave; for example, the Bergeron algorithm can be used to transform the fault voltage mixed wave signal... Mixed wave signal with fault current Extracting the reverse current wave The calculation formula is: .

[0037] The zero-mode current reverse traveling wave signal refers to the current signal obtained by extracting the reverse traveling wave from the zero-mode component obtained after phase-mode transformation of the three-phase transient current; for example, the zero-mode current reverse traveling wave signal can be obtained by using the Begeron algorithm based on the zero-mode component at node N5. The line-mode current reverse traveling wave signal refers to the average line-mode current reverse traveling wave signal obtained by extracting the reverse traveling waves from the first and second line-mode components obtained after phase-mode transformation of the three-phase transient current, and then averaging the two line-mode current reverse traveling waves; for example, the line-mode current reverse traveling wave signal can be obtained by averaging the first and second line-mode current reverse traveling wave signals at node N5.

[0038] S2. The continuous smooth window width, threshold coefficient, and duration of overthreshold time of the differential dynamic transient operator are globally optimized using the crow search algorithm. In each iteration, a fitness value is calculated based on a set of parameters consisting of the continuous smooth window width, the threshold coefficient, and the duration of overthreshold time. The crow position is updated with the goal of minimizing the fitness value, thereby obtaining the optimal continuous smooth window width, the optimal threshold coefficient, and the optimal duration of overthreshold time.

[0039] The crow search algorithm refers to a metaheuristic optimization algorithm that simulates the behavior of crows following each other and hiding food within a population; for example, assuming a crow population size of... The maximum number of iterations is The probability of perception is The flight step length coefficient is The fitness value is minimized by iteratively updating the crow's position. The differential dynamic transient operator refers to a signal processing operator that detects the arrival time of the wavefront based on the signal's jerk energy; for example, first calculating the jerk energy of the original signal. The jump energy is normalized and smoothed, and then based on the detection threshold... and the proportion of time exceeding the threshold Determine the sampling point of the port. .

[0040] Here, the continuous smoothing window width refers to the time width parameter used to construct the Gaussian smoothing kernel when performing Gaussian smoothing on the quiescent energy; for example, the continuous smoothing window width. by The threshold coefficient is obtained through iterative optimization using the Raven Search algorithm, with units of [value missing]. The threshold coefficient refers to the proportionality coefficient used to calculate the detection threshold from the smoothed energy curve; for example, the threshold coefficient... With smooth energy curve peak Multiply to obtain the detection threshold The duration of sustained over-threshold time is obtained through iterative optimization using the Raven Search algorithm. This duration refers to the minimum time required for the energy curve to continuously exceed the detection threshold when determining the arrival sampling point; for example, the duration of sustained over-threshold time... by The unit is obtained through iterative optimization using the crow search algorithm, and is related to the sampling frequency. Jointly determine the number of sampling points that continuously exceed the threshold .

[0041] The fitness value refers to a quantitative indicator that measures the extraction quality at time of arrival under a set of parameters; a smaller value indicates stronger extraction capability. For example, the fitness value... ,in For the detection threshold, To smooth the energy curve, For the sampling point of the port.

[0042] The optimal continuous smooth window width refers to the continuous smooth window width that minimizes the fitness value, obtained through global optimization using the Raven Search algorithm. For example, in the simulation example conducted on January 1, 2026, the optimal continuous smooth window width obtained after iteration using the Raven Search algorithm was used to construct the Gaussian smoothing kernel. The optimal threshold coefficient refers to the threshold coefficient that minimizes the fitness value, obtained through global optimization using the Raven Search algorithm. For example, in the simulation example conducted on January 1, 2026, the optimal threshold coefficient obtained after iteration using the Raven Search algorithm was used to determine the detection threshold. The optimal sustained overthreshold time refers to the sustained overthreshold time that minimizes the fitness value, obtained through global optimization using the Raven Search algorithm. For example, in the simulation example conducted on January 1, 2026, the optimal sustained overthreshold time obtained after iteration using the Raven Search algorithm was used to calculate the number of sustained overthreshold sampling points.

[0043] S3. Substitute the optimal continuous smooth window width, the optimal threshold coefficient, and the optimal sustained overthreshold time into the differential dynamic transient operator, and extract the arrival sampling points of the zero-mode current reverse traveling wave signal and the line-mode current reverse traveling wave signal respectively. Calculate the time difference between the zero-mode arrival time and the line-mode arrival time of the monitoring point based on the zero-mode arrival sampling points and line-mode arrival sampling points extracted for each monitoring point. Construct a time difference vector based on the time differences of all monitoring points, and determine the monitoring point corresponding to the smallest element in the time difference vector as the reference node.

[0044] Here, the arrival sampling point refers to the discrete sampling point number corresponding to the first arrival of the traveling wave signal at the monitoring point; for example, when the over-threshold duration ratio is satisfied. When the threshold is exceeded, take the first one. The sampling points are used as the arrival sampling points. Zero-mode arrival sampling points refer to the arrival sampling points obtained from the zero-mode current reverse traveling wave signal; for example, the zero-mode arrival sampling point number is obtained by applying optimal parameters to the zero-mode current reverse traveling wave signal at node N5. Line-mode arrival sampling points refer to the arrival sampling points obtained from the reverse traveling wave signal of the line-mode current; for example, the line-mode arrival sampling point number is obtained by applying optimal parameters to the reverse traveling wave signal of the line-mode current at node N5. .

[0045] Here, zero-mode arrival time refers to the actual time corresponding to the zero-mode arrival sampling point; for example, the zero-mode arrival sampling point sequence number. With sampling frequency The ratio is the zero-mode arrival time. In this embodiment, the zero-mode arrival time of node N5 is... Linear mode arrival time refers to the actual time corresponding to the line mode arrival sampling point; for example, the line mode arrival sampling point number. With sampling frequency The ratio is the linear mode arrival time. In the embodiment, the linear mode arrival time of node N5 is... The time difference vector refers to a one-dimensional vector consisting of the time difference between the zero-mode arrival time and the line-mode arrival time at each monitoring point; for example, in this embodiment, the time difference vector for the five monitoring points is... .

[0046] Here, the reference node refers to the monitoring point corresponding to the smallest element in the time difference vector; for example, the time difference of node N5 in the embodiment. Since it is the smallest, node N5 is determined as the reference node.

[0047] S4. Obtain the pre-fitted nonlinear relationship between zero-mode wave velocity and fault distance, construct a two-end ranging equation with the reference node and each auxiliary node respectively, and solve the fault distance estimate corresponding to each auxiliary node according to the nonlinear relationship and the two-end ranging equation.

[0048] Here, the nonlinear relationship refers to a fractional function expression describing the variation of zero-mode wave velocity with fault distance; for example, the fitted relationship between zero-mode wave velocity and fault distance is: ,in , , , The constant parameters are obtained through pre-fitting; the fitting result in the example is... Auxiliary nodes refer to monitoring points other than the reference node; for example, in this embodiment, the reference node is N5, and the auxiliary nodes are N1, N2, N3, and N4.

[0049] Among them, the two-end ranging equation refers to an equation constructed based on the ratio of the modulus time difference between the reference node and the auxiliary node, the ratio of the fault distance, and the relationship between the linear mode wave velocity and the zero mode wave velocity; for example, the equation has the following form: This is used to solve for the fault distance estimate. The fault distance estimate refers to the distance from the fault point to the reference node, calculated using the two-terminal ranging equation of the reference node and an auxiliary node. For example, in the embodiment, the fault distance estimate vectors calculated by the reference node N5 and the auxiliary nodes N1, N2, N3, and N4 are as follows: .

[0050] S5. Perform a weighted average on the estimated fault distance to obtain the final fault distance.

[0051] The final fault distance refers to the final location result obtained by weighted averaging of all fault distance estimates. For example, in this embodiment, the reciprocal of the distance from reference node N5 to each auxiliary node is used as the weight, and the final fault distance is obtained after weighted averaging. .

[0052] The technical solution of this embodiment collects three-phase transient currents and three-phase transient voltages at various monitoring points in the distribution network and performs phase-mode transformation to extract zero-mode and line-mode current reverse traveling wave signals. It uses a crow search algorithm to globally optimize the continuous smoothing window width, threshold coefficient, and duration of over-threshold time of the differential dynamic transient operator to obtain the optimal parameter combination. The optimal parameters are then substituted into the differential dynamic transient operator to extract the zero-mode and line-mode wave arrival sampling points at each monitoring point and calculate the time difference to determine the reference node. A two-end ranging equation is constructed by combining a pre-fitted nonlinear relationship between zero-mode wave velocity and fault distance to solve the fault distance estimate and perform a weighted average. This solves the problems of excessive dead zones and ranging errors caused by transition resistance, load fluctuations, and changes in operating modes in multi-branch dominated power grids, as well as the insufficient accuracy and robustness of wavehead extraction caused by reflected wave aliasing, wavehead identification difficulties, wave velocity uncertainty, and strong noise interference in multi-branch topologies of the traditional impedance method. This improves the accuracy and anti-interference capability of fault ranging in complex distribution networks.

[0053] In one alternative approach, the steps of performing phase-mode transformation based on the three-phase transient currents and the three-phase transient voltages and extracting the reverse traveling wave to obtain the zero-mode current reverse traveling wave signal and the line-mode current reverse traveling wave signal include: The three-phase transient current and the three-phase transient voltage are subjected to phase-mode transformation using the Karrenbauer transformation matrix, which is: ;in, This represents the phase-mode transformation matrix, used to transform the three-phase transient current and three-phase transient voltage into zero-mode components, first line-mode components, and second line-mode components, respectively.

[0054] The Begeron algorithm is used to extract the inverse traveling wave from the zero-mode component, the first line-mode component, and the second line-mode component, respectively; wherein, the Begeron algorithm is based on the fault voltage mixed wave signal. Fault current mixed wave signal and line impedance Calculate the reverse traveling wave of the current The calculation formula is: .

[0055] The zero-mode current reverse traveling wave signal is obtained based on the zero-mode component, the first line-mode current reverse traveling wave signal is obtained based on the first line-mode component, and the second line-mode current reverse traveling wave signal is obtained based on the second line-mode component; the first line-mode current reverse traveling wave signal and the second line-mode current reverse traveling wave signal are averaged to obtain an average line-mode current reverse traveling wave signal, which is used as the line-mode current reverse traveling wave signal.

[0056] In the above-mentioned optional methods, the three-phase transient current and voltage are further transformed by the Karrenbauer transformation matrix to separate the zero-mode component and two line-mode components. The Begeron algorithm is used to extract the reverse traveling wave of each mode current. The two line-mode current reverse traveling waves are averaged to obtain the average line-mode current reverse traveling wave signal, which improves the accuracy of mode separation and the stability of reverse traveling wave extraction, and provides a balanced line-mode signal basis for subsequent wave arrival time detection.

[0057] In one alternative approach, the steps for calculating the fitness value include: The jog energy is calculated based on the zero-mode current reverse traveling wave signal and the line-mode current reverse traveling wave signal. Regarding the jump energy Normalization is performed to obtain normalized energy. The normalized energy is obtained by using a Gaussian smoothing kernel. Convolution smoothing yields a smoothed energy curve. According to the threshold coefficient With the smooth energy curve peak Determine the detection threshold This will be the first time that the threshold for sustained proportion is met. The sampling point corresponding to the time is used as the arrival sampling point. The duration of exceeding the threshold is: ;in, This is an indicator function that takes the value 1 when the condition is true and 0 otherwise.

[0058] fitness value The calculation formula is: .

[0059] It should be noted that, This represents a parameter vector consisting of a continuously smooth window width, a threshold coefficient, and a sustained over-threshold time.

[0060] In the above-mentioned optional methods, the jump energy is further calculated based on the reverse traveling wave of the zero-mode and line-mode currents and normalized. The normalized energy is then convolved with a Gaussian smoothing kernel to obtain a smooth energy curve. The detection threshold is determined based on the threshold coefficient and the peak value. The arrival sampling point is determined by combining the over-threshold duration ratio. The detection quality is measured by the fitness function, which improves the wavefront identification accuracy and the reliability of wave arrival time extraction.

[0061] In one alternative approach, the Gaussian smoothing kernel is constructed as follows: based on the sampling frequency Continuous smooth window width Converting from time scale to sampling point scale ,Will The standard deviation of the Gaussian smoothing kernel is calculated using the full width at half maximum (FWHM). In discrete networks Construct a normalized Gaussian smoothing kernel: in, To smooth the kernel half-width, the value is [value to be filled in]. .

[0062] In the above-mentioned optional methods, the continuous smoothing window width is further converted from the time scale to the sampling point scale according to the sampling frequency. The standard deviation of the Gaussian smoothing kernel is calculated using the full width at half maximum (FWHM). A normalized Gaussian smoothing kernel is constructed on the discrete network and the half width of the smoothing kernel is determined. This ensures the adaptability of the smoothing kernel parameters to the sampling frequency and improves the continuity and boundary accuracy of the energy curve smoothing process.

[0063] In one alternative approach, the number of continuously exceeding the threshold sampling points According to the duration of the overthreshold and sampling frequency Confirmed, the calculation formula is as follows: .

[0064] In the above-mentioned optional methods, the number of continuous over-threshold sampling points is further determined based on the continuous over-threshold time and the sampling frequency. The time parameter is directly converted into the number of discrete sampling points, so that the over-threshold duration ratio determination strictly corresponds to the digital sampling sequence, which enhances the compatibility between the continuous over-threshold determination and the sampling process and improves the timeliness consistency of the arrival sampling point detection.

[0065] In one alternative approach, when performing global optimization using a crow search algorithm, the crow population size is assumed to be... The maximum number of iterations is The probability of perception is The flight step length coefficient is .

[0066] In each iteration, for the th One crow, randomly select another different crow Generate random numbers that follow a uniform distribution. ;like Then the first The location of the crow has been updated to ,in For crows The historical best memory location; if Then the first The location of the crow has been updated to ,in and These are the lower and upper bounds of the search space, respectively.

[0067] Among the above optional methods, the population size, maximum number of iterations, perception probability and flight step size coefficient are further set. Based on the relationship between the random number and the perception probability, the crow position is updated by following the historical best memory position or by random search. The goal is to minimize the fitness value for global optimization, which improves the convergence speed and global search capability of the parameter optimization process.

[0068] In one alternative approach, the nonlinear relationship between the zero-mode wave velocity and the fault distance is a fractional function: in, For monitoring points Zero-mode wave velocity at that location From the fault point to the monitoring point distance, , , , These are the constant parameters obtained through pre-fitting.

[0069] In the above-mentioned optional methods, a fractional function is further used to fit the nonlinear relationship between zero-mode wave velocity and fault distance. The variation law of zero-mode wave velocity under different fault distances is described by pre-calibrated constant parameters. The nonlinear relationship is substituted into the two-end ranging equation, which improves the accuracy of zero-mode wave velocity value and provides a reliable wave velocity basis for fault distance calculation.

[0070] In one alternative approach, when performing a weighted average of the fault distance estimates, the fault distance estimate for each auxiliary node is... weight Reference node To the auxiliary node distance The reciprocal of, that is .

[0071] The weights are normalized to obtain normalized weights. ,in The total number of auxiliary nodes; the final fault distance is: .

[0072] In the above-mentioned optional methods, the inverse of the distance from the reference node to the auxiliary node is further used as the weight of the fault distance estimate of each auxiliary node. After normalizing the weights, a weighted average is calculated, which reduces the interference of the estimation error of the long-distance auxiliary node on the final result and improves the accuracy of the comprehensive fault distance estimation and the rationality of the contribution of each node.

[0073] In one alternative approach, the two-end ranging equation constructed by the reference node and each auxiliary node is: in, Reference node The time difference between the zero-mode wave arrival time and the line-mode wave arrival time. as auxiliary node The time difference mentioned above, The reference node From the fault point to the auxiliary node The shortest distance, From the fault point to the reference node distance, For linear mode wave velocity, and Determined based on the aforementioned nonlinear relationship.

[0074] Among the above-mentioned optional methods, a multivariate two-end ranging equation that includes time difference ratio, fault distance ratio, and the relationship between line mode and zero mode wave velocity is further constructed. This fully utilizes the difference in zero mode arrival time and nonlinear wave velocity relationship between the reference node and the auxiliary node, thereby improving the rigor of multi-end information fusion ranging and enhancing the accuracy of fault distance calculation under complex topology structures.

[0075] In another embodiment of the CSA-DDT-optimized distribution network fault location method of the present invention, the method is applied to a typical radial distribution network system, where all lines are uniform, lossless overhead lines. For example... Figure 2 As shown, the distribution network includes a central bus N6 and five line terminals N1, N2, N3, N4, and N5. Fault point F is located on the line segment between N5 and N6, with a distance of 4.5 km between fault point F and node N5. Traveling wave monitoring devices are installed at each line terminal, i.e., nodes N1, N2, N3, N4, and N5, to synchronously collect transient fault signals.

[0076] Real-time monitoring and acquisition of the three-phase transient current of the traveling wave monitoring device at each monitoring point. , , and three-phase transient voltage , , ,in Indicates the monitoring point number. Representing time. Phase-mode transformation is performed on the fault three-phase current and fault three-phase voltage collected at each monitoring point. The Karrenbauer transformation matrix is ​​used: The three-phase transient current and three-phase transient voltage are transformed into zero-mode components, first-line-mode components, and second-line-mode components. The specific transformation formula is as follows: in , For zero-mode current and voltage, , For the first linear modulus component, , This is the second linear modulus component.

[0077] The Begeron algorithm was used to extract the inverse traveling wave from the zero-mode component, the first linear mode component, and the second linear mode component, respectively. The Begeron algorithm is based on the fault voltage mixed wave signal. Fault current mixed wave signal and line impedance Calculate the reverse traveling wave of the current The calculation formula is: The derivation process is as follows: from the traveling wave and anti-traveling wave satisfy , Solving the system of equations yields the following results: Then, according to Ohm's law, In this embodiment, the zero-mode impedance The value is 806.3Ω, which is the line-mode impedance. The value is 329.0Ω. The zero-mode current reverse traveling wave signal is obtained from the zero-mode component, the first line-mode current reverse traveling wave signal is obtained from the first line-mode component, and the second line-mode current reverse traveling wave signal is obtained from the second line-mode component. The first and second line-mode current reverse traveling wave signals are averaged to obtain the average line-mode current reverse traveling wave signal, as shown in the formula: The average line-mode current reverse traveling wave signal is used as the line-mode current reverse traveling wave signal. For example... Figure 3As shown, comparing the anti-traveling wave waveform obtained at node N5 with the mixed wave waveform, the wavefront of the anti-traveling wave is steeper and clearer than that of the mixed wave. Figure 3 The amplitude variation curves of zero-mode mixed wave, zero-mode inverse traveling wave, linear-mode mixed wave, and linear-mode inverse traveling wave over time are given respectively.

[0078] Using the zero-mode current reverse traveling wave signal and the line-mode current reverse traveling wave signal as the original signals, the Raven search algorithm is applied to the continuous smoothing window width of the differential dynamic transient operator. Threshold coefficient and duration of overthreshold Perform global optimization.

[0079] First, define the jerk energy. For the original signal... sampling frequency Set to 10MHz, time interval: Calculation speed: Calculate acceleration: The jump energy is: .

[0080] The jerk energy obtained from the zero-mode and line-mode signals at each monitoring point is normalized to obtain the normalized energy: ;in This represents the peak value of the signal's judder energy.

[0081] A smoothed energy curve is obtained by convolution smoothing the normalized energy using a Gaussian smoothing kernel. The Gaussian smoothing kernel is constructed as follows: based on the sampling frequency... Continuous smooth window width Converting from time scale to sampling point scale: in The unit is μs. As the full width at half maximum (FWHM), calculate the standard deviation of the Gaussian smoothing kernel: In discrete networks Construct a normalized Gaussian smoothing kernel: in The kernel half-width is smoothed. The energy curve is smoothed using convolution: Based on the threshold coefficient With smooth energy curve peak Determine the detection threshold: Based on the duration of overthreshold and sampling frequency Determine the number of sampling points that continuously exceed the threshold: The duration of exceeding the threshold is defined as: in This is an indicator function that takes the value 1 when the condition is true and 0 otherwise. When the first sampling point that exceeds the detection threshold is selected, it is taken as the arrival sampling point. .

[0082] fitness value The calculation formula is: The parameter vector is processed using the Raven Search algorithm. Perform a global optimization. Set the crow population size to [value missing]. The maximum number of iterations is The probability of perception is The flight step length coefficient is Definition of the first The position vector of the crow is a continuous parameter vector to be optimized. Based on the search lower bound LB and search upper bound UB, the initial population position matrix is ​​randomly generated in the solution space. In each iteration, for the ... One crow, randomly select another different crow Generate random numbers that follow a uniform distribution. .like , representing crows Unaware of being followed, the Only crows towards crows Fly from the best historical memory position, and update the position to: in For crows The historical best memory location. If , representing crows Realizing they were being followed, the first A crow randomly flies to any location within the search space, and its location is updated to: After each iteration, the fitness value of the new position is calculated. If the new fitness value is smaller, the crow's historical best position is updated, and the global best parameters of the population are updated simultaneously. The maximum number of iterations is reached. The optimization process ends when the optimal continuous parameter combination is reached, and the globally optimal combination of parameters is output. .like Figure 4As shown, the horizontal axis of the crow search algorithm convergence curve represents the number of iterations, and the vertical axis represents the fitness value. The fitness values ​​of the zero modulus and the linear modulus tend to stabilize after about 10 iterations, converging to around -52.4 and -70.0, respectively.

[0083] For each monitoring point, the optimal parameter combination is calculated for both the zero-mode and line-mode signals. These optimal parameters are then substituted into the differential dynamic transient operator, and smoothing, threshold determination, and over-threshold duration ratio discrimination are re-executed to extract the zero-mode arrival sampling points for each monitoring point. and line mode wave sampling points .like Figure 5 As shown in the time-of-arrival process diagram, the top part represents the jump energy of the zero mode. and smooth energy The curve below represents the jump energy of the linear modulus. and smooth energy The curve and arrow indicate the location of the arrival sampling point. The time difference between the zero-mode arrival time and the line-mode arrival time at each monitoring point is: Construct the network's time difference of arrival vector: in This represents the total number of data collection nodes. In this embodiment, the time difference vector for the five monitoring points is: The node with the smallest time difference vector is the one closest to the fault. Node N5 has the smallest time difference of 1.8 μs, therefore it is selected as the reference node. At node N5, the line mode arrival time is 0.025015 s, and the zero mode arrival time is 0.0250168 s. The calculated time difference of arrival... .

[0084] Obtain the pre-fitted nonlinear relationship between zero-mode wave velocity and fault distance. Fit the relationship between zero-mode wave velocity and fault distance using a fractional function to obtain: in For monitoring points Zero-mode wave velocity at that location From the fault point to the monitoring point distance, , , , These are the constant parameters obtained through pre-fitting. In this embodiment, the fitting result is: like Figure 6As shown in the fitted curve of zero-mode wave velocity versus fault distance, the horizontal axis represents the fault distance. (Unit: km), with the vertical axis representing the zero-mode wave velocity. (Unit: m / s) The fitting curve (solid line) of the fractional model matches well with the simulated scatter data (points), verifying the accuracy of the fitting formula.

[0085] Two-end ranging equations are constructed using reference node N5 and each auxiliary node (N1, N2, N3, N4). The modulus time difference between the reference node and the auxiliary nodes satisfies the following relationship: in The time difference between the zero-mode arrival time and the line-mode arrival time at reference node N5 is used as an example. as auxiliary node That time difference, The reference node N5 passes through the fault point to the auxiliary node. The shortest distance, This is the distance from the fault point to the reference node N5. For linear mode wave velocity, and The fault distance vectors are determined based on the aforementioned nonlinear relationships. Substituting the specific parameters and the modulus time difference, we obtain the fault distance vectors corresponding to all ratios. A weighted average is applied to the fault distance estimates. The fault distance estimate for each auxiliary node is... weight From reference node N5 to auxiliary node distance The reciprocal of, that is: The weights of each line are normalized to obtain the normalized weights: in The total number of auxiliary nodes (in this embodiment) The final fault distance is: The calculated final fault distance is 4.505km, which means the distance from fault point F to reference node N5 is 4.505km. The error is very small compared with the actual fault distance of 4.5km, verifying the accuracy of this method.

[0086] This embodiment employs a raven search algorithm to globally optimize the continuous smooth window width, threshold coefficient, and duration of over-threshold for the differential dynamic transient operator. This enables time-of-arrival (TOA) detection to adapt to different noise environments and line conditions, completely avoiding subjective biases caused by manually setting threshold parameters based on experience. A time difference vector is constructed using the time difference between the zero-mode TOA and the line-mode TOA. This time difference depends only on the internal sampling clock of a single monitoring point, eliminating the need for absolute clock synchronization between different monitoring points and effectively overcoming the excessive reliance on high-precision synchronous clocks in traditional traveling wave methods. A fractional function is used to fit the nonlinear relationship between the zero-mode wave velocity and the fault distance, resolving the inaccuracy of the constant wave velocity assumption caused by wave velocity dispersion. These techniques work synergistically to improve the fault location accuracy and engineering practicality of complex multi-mode controlled power grids under harsh conditions such as high-resistance grounding, strong noise, and wave velocity dispersion.

[0087] Figure 7 This diagram illustrates a structural schematic of an embodiment of a distribution network fault location system 200 optimized based on CSA-DDT provided by the present invention. Figure 7 As shown, the CSA-DDT-optimized distribution network fault location system 200 includes: Processing module 201 is used to collect three-phase transient current and three-phase transient voltage at each monitoring point of the distribution network, perform phase mode transformation based on the three-phase transient current and three-phase transient voltage and extract the reverse traveling wave to obtain zero-mode current reverse traveling wave signal and line-mode current reverse traveling wave signal; The iteration module 202 is used to globally optimize the continuous smooth window width, threshold coefficient and continuous overthreshold time of the differential dynamic transient operator using the crow search algorithm. In each iteration process, the fitness value is calculated based on a set of parameters consisting of the continuous smooth window width, the threshold coefficient and the continuous overthreshold time. The crow position is updated with the goal of minimizing the fitness value, so as to obtain the optimal continuous smooth window width, the optimal threshold coefficient and the optimal continuous overthreshold time. The determination module 203 is used to substitute the optimal continuous smooth window width, the optimal threshold coefficient, and the optimal sustained overthreshold time into the differential dynamic transient operator, extract the arrival sampling points of the zero-mode current reverse traveling wave signal and the line-mode current reverse traveling wave signal respectively, calculate the time difference between the zero-mode arrival time and the line-mode arrival time of the monitoring point based on the zero-mode arrival sampling points and the line-mode arrival sampling points extracted for each monitoring point, construct a time difference vector based on the time differences of all monitoring points, and determine the monitoring point corresponding to the smallest element in the time difference vector as the reference node; The calculation module 204 is used to obtain the pre-fitted nonlinear relationship between zero-mode wave velocity and fault distance, construct a two-end ranging equation with the reference node and each auxiliary node respectively, and solve the fault distance estimate corresponding to each auxiliary node according to the nonlinear relationship and the two-end ranging equation. The output module 205 is used to perform weighted averaging on the estimated fault distance to obtain the final fault distance.

[0088] In one alternative embodiment, the processing module 201 is specifically used for: The three-phase transient current and the three-phase transient voltage are subjected to phase-mode transformation using the Karrenbauer transformation matrix, which is: ;in, This represents the phase-mode transformation matrix, used to transform the three-phase transient current and three-phase transient voltage into zero-mode components, first line-mode components, and second line-mode components, respectively. The Begeron algorithm is used to extract the inverse traveling wave from the zero-mode component, the first line-mode component, and the second line-mode component, respectively; wherein, the Begeron algorithm is based on the fault voltage mixed wave signal. Fault current mixed wave signal and line impedance Calculate the reverse traveling wave of the current The calculation formula is: ; The zero-mode current reverse traveling wave signal is obtained based on the zero-mode component, the first line-mode current reverse traveling wave signal is obtained based on the first line-mode component, and the second line-mode current reverse traveling wave signal is obtained based on the second line-mode component; the first line-mode current reverse traveling wave signal and the second line-mode current reverse traveling wave signal are averaged to obtain an average line-mode current reverse traveling wave signal, which is used as the line-mode current reverse traveling wave signal.

[0089] In one alternative approach, the steps for calculating the fitness value include: The jog energy is calculated based on the zero-mode current reverse traveling wave signal and the line-mode current reverse traveling wave signal. Regarding the jump energy Normalization is performed to obtain normalized energy. The normalized energy is obtained by using a Gaussian smoothing kernel. Convolution smoothing yields a smoothed energy curve. According to the threshold coefficient With the smooth energy curve peak Determine the detection threshold This will be the first time that the threshold for sustained proportion is met. The sampling point corresponding to the time is used as the arrival sampling point. The duration of exceeding the threshold is: ;in, This is an indicator function that takes the value 1 when the condition is true and 0 otherwise. fitness value The calculation formula is: .

[0090] In one alternative approach, the Gaussian smoothing kernel is constructed as follows: based on the sampling frequency Continuous smooth window width Converting from time scale to sampling point scale ,Will The standard deviation of the Gaussian smoothing kernel is calculated using the full width at half maximum (FWHM). In discrete networks Construct a normalized Gaussian smoothing kernel: in, To smooth the kernel half-width, the value is [value to be filled in]. .

[0091] In one alternative approach, the number of continuously exceeding the threshold sampling points According to the duration of the overthreshold and sampling frequency Confirmed, the calculation formula is as follows: .

[0092] In one alternative approach, when performing global optimization using a crow search algorithm, the crow population size is assumed to be... The maximum number of iterations is The probability of perception is The flight step length coefficient is ; In each iteration, for the th One crow, randomly select another different crow Generate random numbers that follow a uniform distribution. ;like Then the first The location of the crow has been updated to ,in For crows The historical best memory location; if Then the first The location of the crow has been updated to ,in and These are the lower and upper bounds of the search space, respectively.

[0093] In one alternative approach, the nonlinear relationship between the zero-mode wave velocity and the fault distance is a fractional function: in, For monitoring points Zero-mode wave velocity at that location From the fault point to the monitoring point distance, , , , These are the constant parameters obtained through pre-fitting.

[0094] In one alternative approach, when performing a weighted average of the fault distance estimates, the fault distance estimate for each auxiliary node is... weight Reference node To the auxiliary node distance The reciprocal of, that is ; The weights are normalized to obtain normalized weights. ,in The total number of auxiliary nodes; the final fault distance is: .

[0095] In one alternative approach, the two-end ranging equation constructed by the reference node and each auxiliary node is: in, Reference node The time difference between the zero-mode wave arrival time and the line-mode wave arrival time. as auxiliary node The time difference mentioned above, The reference node From the fault point to the auxiliary node The shortest distance, From the fault point to the reference node distance, For linear mode wave velocity, and Determined based on the aforementioned nonlinear relationship.

[0096] It should be noted that the beneficial effects of the CSA-DDT-optimized distribution network fault location system 200 provided in the above embodiments are the same as those of the CSA-DDT-optimized distribution network fault location method described above, and will not be repeated here. Furthermore, the system provided in the above embodiments is only illustrated by the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the system can be divided into different functional modules according to the actual situation to complete all or part of the functions described above. In addition, the system and method embodiments provided in the above embodiments belong to the same concept, and their specific implementation process is detailed in the method embodiments, and will not be repeated here.

[0097] The fault location system 200 of the distribution network based on CSA-DDT optimization of the present invention can be a computer program (including program code) running on a computer device. For example, the fault location system 200 of the distribution network based on CSA-DDT optimization of the present invention is an application software that can be used to execute the corresponding steps in the fault location method of the distribution network based on CSA-DDT optimization of the present invention.

[0098] In some embodiments, the CSA-DDT-optimized distribution network fault location system 200 of the present invention can be implemented in a combination of hardware and software. As an example, the CSA-DDT-optimized distribution network fault location system 200 of the present invention can be a processor in the form of a hardware decoding processor, which is programmed to execute the CSA-DDT-optimized distribution network fault location method of the present invention. For example, the processor in the form of a hardware decoding processor can be one or more application-specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), complex programmable logic devices (CPLDs), field-programmable gate arrays (FPGAs), or other electronic components.

[0099] The modules described in the embodiments of this invention can be implemented in software or hardware. The names of the modules are not, in some cases, limiting the scope of the module itself.

[0100] An electronic device according to an embodiment of the present invention includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements any of the above-mentioned distribution network fault location methods based on CSA-DDT optimization. That is, an electronic device according to an embodiment of the present invention may include, but is not limited to: a processor and a memory; the memory is used to store the computer program; the processor is used to execute the distribution network fault location method based on CSA-DDT optimization shown in any embodiment of the present invention by calling the computer program.

[0101] In one alternative embodiment, an electronic device is provided, such as Figure 8 As shown, Figure 8The illustrated electronic device 4000 includes a processor 4001 and a memory 4003. The processor 4001 and the memory 4003 are connected, for example, via a bus 4002. Optionally, the electronic device 4000 may further include a transceiver 4004, which can be used for data interaction between the electronic device and other electronic devices, such as sending and / or receiving data. It should be noted that in practical applications, the transceiver 4004 is not limited to one type, and the structure of the electronic device 4000 does not constitute a limitation on the embodiments of the present invention.

[0102] Processor 4001 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this invention. Processor 4001 may also be a combination that implements computational functions, such as including one or more microprocessor combinations, a combination of a DSP and a microprocessor, etc.

[0103] Bus 4002 may include a path for transmitting information between the aforementioned components. Bus 4002 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. Bus 4002 can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 8 The bus 4002 is represented by only one thick line, but this does not mean that there is only one bus or one type of bus.

[0104] The memory 4003 may be ROM (Read Only Memory) or other types of static storage devices capable of storing static information and instructions, RAM (Random Access Memory) or other types of dynamic storage devices capable of storing information and instructions, or EEPROM (Electrically Erasable Programmable Read Only Memory), CD-ROM (Compact Disc Read Only Memory) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto.

[0105] The memory 4003 stores application code (computer program) for executing the present invention, and its execution is controlled by the processor 4001. The processor 4001 executes the application code stored in the memory 4003 to implement the content shown in the foregoing method embodiments.

[0106] Among them, electronic devices can also be terminal devices. A terminal device can be any terminal device that can install applications and access web pages through applications, including at least one of smartphones, tablets, laptops, desktop computers, smart speakers, smartwatches, smart TVs, and smart in-vehicle devices.

[0107] It should be noted that, Figure 8 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of use of the embodiments of the present invention.

[0108] An embodiment of the present invention provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements any of the above-mentioned distribution network fault location methods based on CSA-DDT optimization.

[0109] Alternatively, the computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a compact disc read-only memory (CD-ROM), magnetic tape, a floppy disk, and an optical data storage device, etc.

[0110] In an exemplary embodiment, a computer program product or computer program is also provided, which includes computer instructions stored in a computer-readable storage medium. A processor of an electronic device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the electronic device to perform the aforementioned CSA-DDT-optimized distribution network fault location method.

[0111] Computer program code for performing the operations of this invention can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0112] It should be understood that the flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of methods and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0113] The computer-readable storage medium provided in this invention can be, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0114] The aforementioned computer-readable storage medium carries one or more programs, which, when executed by the electronic device, cause the electronic device to perform the method shown in the above embodiments.

[0115] The above description is merely a preferred embodiment of the present invention and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of disclosure in this invention is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-disclosed concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features with similar functions disclosed in this invention.

[0116] It should be noted that the terms "first," "second," etc., used in the specification and claims of this application are used to distinguish similar objects and represent a limitation on a specific order or sequence. Where appropriate, the order of use for similar objects can be interchanged so that the embodiments of this application described herein can be implemented in an order other than that shown or described.

[0117] Those skilled in the art will recognize that this invention can be implemented as a system, method, or computer program product. Therefore, this invention can be specifically implemented in the following forms: it can be entirely hardware, entirely software (including firmware, resident software, microcode, etc.), or a combination of hardware and software, generally referred to herein as a "circuit," "module," or "system." Furthermore, in some embodiments, this invention can also be implemented as a computer program product contained in one or more computer-readable media, which includes computer-readable program code.

[0118] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A distribution network fault location method based on CSA-DDT optimization, characterized in that, include: The three-phase transient current and three-phase transient voltage at each monitoring point of the distribution network are collected. Based on the three-phase transient current and three-phase transient voltage, phase mode transformation is performed and reverse traveling waves are extracted to obtain zero-mode current reverse traveling wave signals and line-mode current reverse traveling wave signals. The crow search algorithm is used to globally optimize the continuous smooth window width, threshold coefficient, and duration of overthreshold for the differential dynamic transient operator. In each iteration, the fitness value is calculated based on a set of parameters consisting of the continuous smooth window width, the threshold coefficient, and the duration of overthreshold. The crow position is updated with the goal of minimizing the fitness value, so as to obtain the optimal continuous smooth window width, the optimal threshold coefficient, and the optimal duration of overthreshold. Substituting the optimal continuous smooth window width, the optimal threshold coefficient, and the optimal sustained overthreshold time into the differential dynamic transient operator, the arrival sampling points of the zero-mode current reverse traveling wave signal and the line-mode current reverse traveling wave signal are extracted respectively. Based on the zero-mode arrival sampling points and line-mode arrival sampling points extracted for each monitoring point, the time difference between the zero-mode arrival time and the line-mode arrival time of the monitoring point is calculated. A time difference vector is constructed based on the time differences of all monitoring points, and the monitoring point corresponding to the smallest element in the time difference vector is determined as the reference node. Obtain the pre-fitted nonlinear relationship between zero-mode wave velocity and fault distance, construct a two-end ranging equation with the reference node and each auxiliary node respectively, and solve the fault distance estimate corresponding to each auxiliary node according to the nonlinear relationship and the two-end ranging equation. The estimated fault distance is weighted and averaged to obtain the final fault distance.

2. The distribution network fault location method based on CSA-DDT optimization according to claim 1, characterized in that, The steps of performing phase-mode transformation and extracting the reverse traveling wave based on the three-phase transient current and the three-phase transient voltage to obtain the zero-mode current reverse traveling wave signal and the line-mode current reverse traveling wave signal include: The three-phase transient current and the three-phase transient voltage are subjected to phase-mode transformation using the Karrenbauer transformation matrix, which is: ;in, This represents the phase-mode transformation matrix, used to transform the three-phase transient current and three-phase transient voltage into zero-mode components, first line-mode components, and second line-mode components, respectively. The Begeron algorithm is used to extract the inverse traveling wave from the zero-mode component, the first line-mode component, and the second line-mode component, respectively; wherein, the Begeron algorithm is based on the fault voltage mixed wave signal. Fault current mixed wave signal and line impedance Calculate the reverse traveling wave of the current The calculation formula is: ; The zero-mode current reverse traveling wave signal is obtained based on the zero-mode component, the first line-mode current reverse traveling wave signal is obtained based on the first line-mode component, and the second line-mode current reverse traveling wave signal is obtained based on the second line-mode component; the first line-mode current reverse traveling wave signal and the second line-mode current reverse traveling wave signal are averaged to obtain an average line-mode current reverse traveling wave signal, which is used as the line-mode current reverse traveling wave signal.

3. The distribution network fault location method based on CSA-DDT optimization according to claim 1, characterized in that, The steps for calculating the fitness value include: The jog energy is calculated based on the zero-mode current reverse traveling wave signal and the line-mode current reverse traveling wave signal. Regarding the jump energy Normalization is performed to obtain normalized energy. The normalized energy is obtained by using a Gaussian smoothing kernel. Convolution smoothing yields a smoothed energy curve. According to the threshold coefficient With the smooth energy curve peak Determine the detection threshold This will be the first time that the threshold for sustained proportion is met. The sampling point corresponding to the time is used as the arrival sampling point. The duration of exceeding the threshold is: ;in, This is an indicator function that takes the value 1 when the condition is true and 0 otherwise. fitness value The calculation formula is: .

4. The distribution network fault location method based on CSA-DDT optimization according to claim 3, characterized in that, The Gaussian smoothing kernel is constructed as follows: based on the sampling frequency Continuous smooth window width Converting from time scale to sampling point scale ,Will The standard deviation of the Gaussian smoothing kernel is calculated using the full width at half maximum (FWHM). In discrete networks Construct a normalized Gaussian smoothing kernel: in, To smooth the kernel half-width, the value is [value to be filled in]. .

5. The distribution network fault location method based on CSA-DDT optimization according to claim 3, characterized in that, The number of continuous over-threshold sampling points According to the duration of the overthreshold and sampling frequency Confirmed, the calculation formula is as follows: .

6. The distribution network fault location method based on CSA-DDT optimization according to claim 1, characterized in that, When performing global optimization using the crow search algorithm, let the crow population size be... The maximum number of iterations is The probability of perception is The flight step length coefficient is ; In each iteration, for the th One crow, randomly select another different crow Generate random numbers that follow a uniform distribution. ;like Then the first The location of the crow has been updated to ,in For crows The historical best memory location; if Then the first The location of the crow has been updated to ,in and These are the lower and upper bounds of the search space, respectively.

7. The distribution network fault location method based on CSA-DDT optimization according to claim 1, characterized in that, The nonlinear relationship between the zero-mode wave velocity and the fault distance is a fractional function: in, For monitoring points Zero-mode wave velocity at that location From the fault point to the monitoring point distance, , , , These are the constant parameters obtained through pre-fitting.

8. The distribution network fault location method based on CSA-DDT optimization according to claim 1, characterized in that, When performing a weighted average of the fault distance estimates, the fault distance estimate for each auxiliary node is... weight Reference node To the auxiliary node distance The reciprocal of, that is ; The weights are normalized to obtain normalized weights. ,in The total number of auxiliary nodes; the final fault distance is: .

9. A distribution network fault location method based on CSA-DDT optimization according to claim 7, characterized in that, The two-end ranging equation constructed by the reference node and each auxiliary node is as follows: in, Reference node The time difference between the zero-mode wave arrival time and the line-mode wave arrival time. as auxiliary node The time difference mentioned above, The reference node From the fault point to the auxiliary node The shortest distance, From the fault point to the reference node distance, For linear mode wave velocity, and Determined based on the aforementioned nonlinear relationship.

10. A distribution network fault location system optimized based on CSA-DDT, characterized in that, include: The processing module is used to collect three-phase transient current and three-phase transient voltage at each monitoring point of the distribution network, perform phase mode transformation based on the three-phase transient current and three-phase transient voltage, and extract the reverse traveling wave to obtain zero-mode current reverse traveling wave signal and line-mode current reverse traveling wave signal; The iterative module is used to globally optimize the continuous smooth window width, threshold coefficient, and duration of overthreshold time of the differential dynamic transient operator using the crow search algorithm. In each iteration, the fitness value is calculated based on a set of parameters consisting of the continuous smooth window width, the threshold coefficient, and the duration of overthreshold time. The crow position is updated with the goal of minimizing the fitness value, so as to obtain the optimal continuous smooth window width, the optimal threshold coefficient, and the optimal duration of overthreshold time. The determination module is used to substitute the optimal continuous smooth window width, the optimal threshold coefficient, and the optimal sustained overthreshold time into the differential dynamic transient operator, extract the arrival sampling points of the zero-mode current reverse traveling wave signal and the line-mode current reverse traveling wave signal respectively, calculate the time difference between the zero-mode arrival time and the line-mode arrival time of the monitoring point based on the zero-mode arrival sampling points and the line-mode arrival sampling points extracted for each monitoring point, construct a time difference vector based on the time differences of all monitoring points, and determine the monitoring point corresponding to the smallest element in the time difference vector as the reference node; The calculation module is used to obtain the pre-fitted nonlinear relationship between zero-mode wave velocity and fault distance, construct a two-end ranging equation with the reference node and each auxiliary node respectively, and solve the fault distance estimate corresponding to each auxiliary node according to the nonlinear relationship and the two-end ranging equation. The output module is used to perform weighted averaging on the estimated fault distance to obtain the final fault distance.

Images