An emtr cable fault location method based on harvester optimization algorithm

Abstract

The application discloses an EMTR cable fault positioning method based on a harvester optimization algorithm, and belongs to the technical field of cable monitoring, and comprises the following steps: initializing parameters; constructing a cable network topology model; injecting a detection signal into the cable network topology model, propagating the detection signal from a test end of the cable network topology model to a fault point, bouncing the detection signal to the test end, and receiving a reflected signal; performing time reversal processing on the reflected signal, constructing a frequency domain compensation function, using the frequency domain compensation function to correct distortion caused by experiencing attenuation and dispersion in an actual signal channel, introducing a harvester optimization algorithm to iteratively optimize compensation parameters of the frequency domain compensation function, and obtaining an optimized inversion signal; based on the optimized inversion signal, obtaining a high PSR candidate peak value, backtracking to an original received signal, calculating local energy in a first traveling wave theoretical time window of a candidate fault point, and selecting a candidate point with the maximum energy as a real fault point. The method solves the problems of main peak ambiguity and large positioning error.

Description

Technical Field

[0001] This application relates to the field of cable monitoring technology, and in particular to an EMTR cable fault location method based on a dung beetle optimization algorithm. Background Technology

[0002] Power cables play a crucial role in power transmission and distribution. During operation, cables are inevitably subjected to the combined effects of electrical, thermal, and mechanical stresses, leading to insulation aging and defects. These aging and defects continue to develop, eventually evolving into more serious cable grounding faults, causing severe accidents such as cable explosions. Research indicates that intermittent grounding arc faults, known as early-stage cable faults, occur before permanent grounding faults. If these early-stage faults are not detected and addressed promptly, the frequency of intermittent arcing and insulation deterioration mutually reinforce each other, ultimately leading to permanent faults. Therefore, accurately locating cable defects during early-stage faults is of great significance for improving the reliability of the power grid.

[0003] Existing technologies employ the traditional EMTR principle for cable fault location. The core of the traditional EMTR principle lies in achieving fault location through signal reverse propagation. The specific implementation process is as follows: First, a pulsed or sinusoidal probe signal is injected into the cable from the test end. When this signal propagates along the cable to the fault point, it experiences a difference in reflection due to impedance abrupt changes. Then, the receiving end captures the reflected signal containing fault characteristic information. At this point, time-reversal processing (taking the complex conjugate of the signal in the frequency domain) is performed on the received signal, an operation that reverses the signal propagation direction. Next, the reversed signal is re-injected into the cable. Utilizing the reversibility of electromagnetic wave propagation, the signal energy is focused at the fault point. Finally, by detecting the arrival time and position of the focused peak, combined with known wave velocity parameters, the physical distance between the fault point and the test end can be accurately calculated. This technology effectively improves the accuracy and anti-interference capability of fault location through signal time symmetry reconstruction.

[0004] However, under complex cable networks (including multiple branches and terminals) and high-resistance or nonlinear fault conditions, traditional EMTR is limited by the following problems: due to the unknown fault impedance, the reflection coefficient model and propagation function model have large deviations, resulting in inaccurate peak localization and focusing after inversion; frequency-varying loss and dispersion cause the forward propagation information to be attenuated and distorted during reverse reconstruction, reducing the robustness of localization; noise and multiple reflections produce false peaks, misjudging the fault location; traditional parameter adjustment relies on experience or exhaustive search, which is inefficient and inaccurate. Summary of the Invention

[0005] To address the aforementioned shortcomings in existing technologies, this application provides an EMTR cable fault location method based on a dung beetle optimization algorithm, which solves the problems of main peak ambiguity and large location error in traditional methods.

[0006] To achieve the aforementioned objectives, the technical solution adopted in this application is as follows:

[0007] This application provides a method for EMTR cable fault location based on a dung beetle optimization algorithm, comprising:

[0008] S1: Construct a cable network topology model for the cable network;

[0009] S2: Inject the detection signal into the cable network topology model, propagate it from the test end of the cable network topology model to the fault point, and then bounce back to the test end to receive the reflected signal containing fault characteristic information;

[0010] S3: Perform time-reversal processing on the reflected signal, construct a frequency domain compensation function, use the frequency domain compensation function to correct the distortion of the inverted signal caused by attenuation and dispersion in the actual signal channel, and introduce the dung beetle optimization algorithm to iteratively optimize the compensation parameters of the frequency domain compensation function to obtain the optimized inverted signal;

[0011] S4: Based on the optimized inversion signal, obtain high PSR candidate peak values, and backtrack to the original received signal to calculate the local energy within the first traveling wave theoretical time window of the candidate fault point. Select the candidate point with the largest energy as the real fault point.

[0012] Furthermore, the construction of a cable network topology model for the cable network includes:

[0013] S101: Abstracting the cable network into a weighted graph model :

[0014]

[0015] In the formula, For a set of nodes, For the test end, As a branch point, The fault point For terminal nodes, For edge set, For nodes To the node cable section, For cable section In frequency The edge weight function below;

[0016] S102: Calculate the propagation constant for each edge. and characteristic impedance ;

[0017] Modeling each segment of the uniform transmission cable as a uniform transmission line, and starting from the telegraph equations, we construct the voltage and current fluctuation equations for the cable line:

[0018]

[0019]

[0020] In the formula, It is a voltage wave. It is a current wave. Angular frequency, For location, The imaginary unit, , , and These are resistance, inductance, conductance, and capacitance per unit length, respectively.

[0021] Solving the wave equation yields the expression for the voltage wave:

[0022]

[0023] In the formula, Forward traveling wave, It is a reverse traveling wave. The propagation constant;

[0024] Solve for the propagation constant based on the expression for voltage waves. and characteristic impedance :

[0025]

[0026]

[0027] In the formula, It is the attenuation constant. It is the phase constant;

[0028] S103: Treat each cable segment as a two-port network and define the scattering parameters. Characterizing the discontinuity in characteristic impedance, which leads to wave reflection and transmission, for Port network, scattering parameters The expression is:

[0029]

[0030] In the formula, For port The normalized incident wave, For port Normalized reflected wave, Indicates from port Voltage waves incident on the network, Indicates from port Voltage waves reflected back into the network For port Characteristic impedance, Indicates from port Incident wave to port Scattering parameters of the emitted wave;

[0031] The scattering parameters of the two-port network are:

[0032]

[0033] In the formula, For series impedance, Characteristic impedance;

[0034] Among them, port Reflectance coefficient for:

[0035]

[0036] In the formula, Indicates from port Incident wave pairs with port The effect of the emitted wave.

[0037] Furthermore, the frequency domain expression of the reflected signal is:

[0038]

[0039] In the formula, The frequency domain representation of the reflected signal. For the frequency domain representation of the injected signal, For the test end propagation to the fault point The forward transfer function, This is the corrected reflection coefficient. and The key parameter for the reflection coefficient, Fault point Reflection propagates to the test end The reverse transfer function, To correct for the effects of noise and stray reflections;

[0040] The total propagation delay is:

[0041]

[0042] In the formula, For the signal at group velocity Along path length The time required for dissemination This is the average operator. For path.

[0043] Further, the reflected signal undergoes time-reversal processing to construct a frequency-domain compensation function. This function is used to correct the distortion of the inverted signal caused by attenuation and dispersion in the actual signal channel. A dung beetle optimization algorithm is introduced to iteratively optimize the compensation parameters of the frequency-domain compensation function, resulting in an optimized inverted signal, including:

[0044] S301: Perform time-reversal processing on the reflected signal and retransmit it;

[0045] S302: Construct a frequency domain compensation function that includes amplitude compensation and phase compensation terms:

[0046]

[0047] In the formula, For frequency domain compensation function, and For amplitude attenuation compensation parameters, and These are the phase delay compensation parameters. The total length of the theoretical propagation path, For the set of compensation parameters;

[0048] S303: In the frequency domain, the received inversion signal is taken as its complex conjugate and multiplied by the compensation function to obtain the corrected inversion signal. Then, for the set of compensation parameters, the compensation parameters of the frequency domain compensation function are iteratively optimized using the dung beetle optimization algorithm to obtain the optimized inversion signal. :

[0049]

[0050] In the formula, The inversion signal obtained by taking the complex conjugate is... , This refers to the phase shift of the signal during propagation;

[0051] S304: Based on The inverse signal in the time domain is obtained using the inverse Fourier transform. :

[0052]

[0053] In the formula, For Fourier transform, superscript This is the inverse operation. For time.

[0054] Furthermore, the iterative optimization of the compensation parameters of the frequency domain compensation function using the dung beetle optimization algorithm includes:

[0055] A1: Initialize the parameters of the dung beetle optimization algorithm and generate them randomly. The initial positions of the candidate solutions are calculated, and the objective function value corresponding to each candidate solution position is calculated. ;

[0056] A2: Update the position based on the current optimal solution and the global optimal solution, with a random step size and direction. The update formula is as follows:

[0057]

[0058] In the formula, For the first The candidate solution at the th... The set of compensation parameters for the next iteration For the first The candidate solution at the th... The set of compensation parameters for the next iteration This is the globally optimal solution. This is the current optimal solution. and This is the step size control factor. It is a random vector. This represents element-wise multiplication;

[0059] A3: To avoid getting trapped in local optima, a small-scale random perturbation is made around the current position to simulate the process of a dung beetle repositioning itself when encountering an obstacle, performing a fine-grained local search. The update formula is:

[0060]

[0061] In the formula, The disturbance intensity factor, A standard normally distributed random vector, It is the identity matrix;

[0062] A4: Retain the current optimal candidate solution and its surrounding region, generate new candidate solutions through Gaussian perturbation, simulate dung beetle reproduction, ensure the inheritance and preservation of excellent individuals, and conduct small-scale exploration. The update formula is:

[0063]

[0064] In the formula, This is the updated set of compensation parameters after Gaussian perturbation. The Gaussian perturbation scaling factor;

[0065] A5: Simulating the behavior of a small dung beetle stealing dung balls from a large dung beetle, a new random exploration direction is introduced to escape local optima. The update formula is:

[0066]

[0067] In the formula, and Step size factor It is a uniformly distributed random vector;

[0068] A6: Repeat A2-A5 until the maximum number of iterations is reached or the convergence condition is met. This yields the optimal parameter vector, where, It is a very small positive number.

[0069] Furthermore, based on the optimized inversion signal, high PSR candidate peak values ​​are obtained, and the original received signal is traced back to calculate the local energy within the first traveling wave theoretical time window of the candidate fault point. The candidate point with the highest energy is selected as the real fault point, including:

[0070] S401: Inversion signal in the time domain obtained using inverse Fourier transform High PSR candidate peak values ​​were obtained:

[0071]

[0072] In the formula, The peak-to-side-lobe ratio, Main petal window This is the side lobe region;

[0073] S402: Time for each candidate peak Calculate the energy of the candidate peak in the inverted signal, and backtrack to the original received signal to calculate the theoretical time window of the first traveling wave at the candidate fault point. Local energy within :

[0074]

[0075] In the formula, The original received signal, It is an extremely small time interval;

[0076] S403: Among all candidate peak values ​​where PSR exceeds the threshold, select the maximum local energy of the corresponding initial traveling wave as the true fault point:

[0077]

[0078] In the formula, This represents the actual point of failure.

[0079] The beneficial effects of this application are:

[0080] This application provides a fault location method for EMTR cables based on the Dung Beetle Optimization (DBO) algorithm. It utilizes a frequency domain compensation function to accurately offset attenuation distortion and phase shift during signal forward propagation, correcting the shortcomings of traditional electromagnetic time-inversion fault location methods, such as "simple complex conjugate inversion," which cannot handle actual channel distortion. Simultaneously, the DBO algorithm is introduced to automatically iteratively optimize compensation parameters, achieving global parameter search and local fine-tuning. This efficiently finds the optimal parameter combination suitable for different fault impedances and channel losses, significantly reducing model bias and greatly improving the focus of the inverted signal at the fault point. This solves the problems of blurred main peaks and large location errors in traditional methods. Attached Figure Description

[0081] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other embodiments can be obtained based on these drawings.

[0082] Figure 1 This is a flowchart illustrating an EMTR cable fault location method based on a dung beetle optimization algorithm, provided in an embodiment of this application.

[0083] Figure 2 This is another flowchart illustrating an EMTR cable fault location method based on a dung beetle optimization algorithm provided in an embodiment of this application.

[0084] Figure 3 This is a flowchart illustrating an embodiment of the present application.

[0085] Figure 4 This is a flowchart illustrating the dung beetle optimization algorithm provided in an embodiment of this application. Detailed Implementation

[0086] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art based on this application are within the scope of protection of this application.

[0087] Terminology Explanation:

[0088] EMTR: Electromagnetic Time Reversal Fault Location Method, is a technique for fault location based on the "time reversal symmetry" of electromagnetic waves.

[0089] DBO: Dung Beetle Optimization Algorithm, is an intelligent algorithm that achieves global / local search and parameter optimization by simulating the behaviors of dung beetles, such as rolling balls, dancing, reproducing, and stealing.

[0090] PSR: Peak-to-Sidelobe Ratio, is an index that measures the ratio of the energy of the main peak of the focused signal to the energy of the sidelobe. A higher value indicates a better focusing effect.

[0091] S-parameters: Generalized Scattering Parameters, matrix parameters used to describe the electromagnetic wave reflection and transmission characteristics at cable network nodes (test ends, branch points, etc.).

[0092] γ: Propagation constant, a parameter describing the attenuation and phase shift characteristics of electromagnetic waves propagating in a cable, γ = α + jβ, where α is the attenuation constant and β is the phase shift constant. It is the imaginary unit.

[0093] Characteristic impedance is the inherent impedance characteristic of a cable, determined by its resistance, inductance, capacitance, and conductance per unit length, and it affects the electromagnetic wave transmission characteristics.

[0094] Traditional EMTR principle:

[0095] 1. Inject a pulse or sine wave detection signal from the test end;

[0096] 2. Receive signals containing fault reflections;

[0097] 3. Perform electromagnetic time inversion on the received signal (taking the complex conjugate in the frequency domain).

[0098] 4. Reinject the inverted signal into the cable, utilizing the reversibility of wave propagation to focus energy on the fault point;

[0099] 5. Calculate the fault distance by identifying the time position of the focusing peak.

[0100] However, traditional EMTR suffers from several drawbacks: unknown or nonlinear fault impedances lead to model biases in inversion based on theoretical reflection coefficients, resulting in mismatch between the amplitude and phase of the focused signal and increased positioning errors; frequency-varying losses and dispersion cannot be fully compensated, so backpropagation cannot completely reconstruct the original waveform, leading to blurred main peaks; noise and multiple reflections result in multiple observable peaks, and simply taking the largest peak can easily lead to misjudgment and false peak interference; manual parameter adjustment is inefficient and not robust, making it difficult to achieve stable positioning in complex scenarios; and fixed parameters reduce the model's versatility and prevent it from addressing more universal situations.

[0101] Based on this, this application provides a fault location method for EMTR cables based on a dung beetle optimization algorithm, which can be found in [reference needed]. Figure 1 and Figure 2 , Figure 1 The diagram shown is a flowchart illustrating an EMTR cable fault location method based on a dung beetle optimization algorithm provided in this application, including:

[0102] S1: Construct a cable network topology model for the cable network.

[0103] Specifically:

[0104] S101: Abstracting the cable network into a weighted graph model .

[0105] In the formula, For a set of nodes, For the test end, As a branch point, The fault point For terminal nodes, For edge set, For nodes To the node cable section, For cable section In frequency The edge weight function under the given conditions characterizes the cable segment at a certain frequency. The transmission characteristics under these conditions, due to their length Characteristic impedance Propagation constant Decide.

[0106] S102: Calculate the propagation constant for each edge. and characteristic impedance .

[0107] like Figure 3As shown, the characteristics of a cable segment are determined by its resistance, inductance, conductance, and capacitance per unit length. Voltage wave propagation is affected by the propagation constant, which determines the signal attenuation constant and phase constant. Characteristic impedance describes the relationship between voltage wave and current and is an important parameter for transmission characteristics.

[0108] Since each segment of a uniform transmission cable can be modeled as a uniform transmission line, and starting from the telegraph equations, we can construct the voltage and current fluctuation equations for the cable:

[0109]

[0110]

[0111] In the formula, It is a voltage wave. It is a current wave. Angular frequency, For location, The imaginary unit, , , and These are resistance, inductance, conductance, and capacitance per unit length, respectively.

[0112] Solving the above equations simultaneously yields the solution for wave propagation. The voltage wave can be decomposed into a forward traveling wave and a reverse traveling wave:

[0113]

[0114] In the formula, Forward traveling wave, It is a reverse traveling wave. The propagation constant is a complex number that determines the changes in amplitude and phase of the wave during propagation. Its expression is:

[0115]

[0116] In the formula, The attenuation constant is the signal attenuation per unit length, expressed in Np / m. is the phase constant, the phase offset per unit length, in rad / m.

[0117] Another key parameter of a transmission line is its characteristic impedance. This represents the ratio of traveling wave voltage to current:

[0118]

[0119] Therefore, the edge weight function The core issue is the length of the cable segment. and its propagation constant and characteristic impedance Defined together.

[0120] S203: At nodes in a cable network (such as branch points, fault points, and terminations), impedance discontinuities can cause wave reflection and transmission, and scattering parameters. It is a standard tool for describing this behavior of multiport networks, for a Port network, its scattering parameters The expression is:

[0121]

[0122] In the formula, For port The normalized incident wave, For port Normalized reflected wave, Indicates from port Voltage waves incident on the network, Indicates from port Voltage waves reflected back into the network For port Characteristic impedance, Indicates from port Incident wave to port Scattering parameters of the emitted wave.

[0123] For a series impedance The fault point can be abstracted as a two-port network as an example. Its S-parameter matrix (assuming the characteristic impedances on both sides are...) )for:

[0124]

[0125] The reflection coefficient of port 1 is ,in, Indicates from port Incident wave pairs with port Scattering parameters of the emitted wave.

[0126] Among them, propagation parameters , Frequency-dependent, it can be calculated using cable material and geometric parameters; , , and Units are , , , Possible port impedance at the node Need and match.

[0127] S2: Inject the detection signal into the cable network topology model, propagate it from the test end of the cable network topology model to the fault point, and then bounce back to the test end to receive the reflected signal containing fault characteristic information.

[0128] In one embodiment of this application, a signal propagates from the test terminal (T) to the fault point (F), and then reflects back to the test terminal (R), forming a cascaded network system. The transmission function from T to F... This is the product of the transmission coefficients of the signal through all cable segments and intermediate nodes. Reflection at point F: This is determined by the corrected reflection coefficient. Function. Transfer function from F to R. , usually with They are highly correlated, even equal in symmetric networks. Therefore, the frequency domain expression of the received signal is:

[0129]

[0130] In the formula, The frequency domain representation of the reflected signal. For the frequency domain representation of the injected signal, For the test end propagation to the fault point The forward transfer function, This is the corrected reflection coefficient. and The key parameter for the reflection coefficient, Fault point Reflection propagates to the test end The reverse transfer function, It is used to correct the effects of noise and stray reflections.

[0131] For a specific path Its total delay The signal is at group velocity Along path length The time required for propagation. Due to dispersion effects, the group velocity is a function of frequency. The average group velocity can be used. Perform an approximate estimation:

[0132]

[0133] In the formula, The signal travels along the path length at group velocity. The time required for dissemination Phase constant , This is the average operator. For path.

[0134] S3: Perform time-reversal processing on the reflected signal, construct a frequency domain compensation function, and use the frequency domain compensation function to correct the distortion of the inverted signal caused by attenuation and dispersion in the actual signal channel. Introduce the dung beetle optimization algorithm to iteratively optimize the compensation parameters of the frequency domain compensation function to obtain the optimized inverted signal.

[0135] In one embodiment of this application, in a real-world channel, signals undergo attenuation and dispersion, resulting in signal distortion. Frequency domain compensation functions are used to compensate for these distortions, including amplitude compensation and phase compensation, with the aim of restoring ideal signal transmission through compensation. These compensation functions use exponential functions to correct the signal.

[0136] The core principle of Electromagnetic Time Reversal (EMTR) is time-reversal invariance. In a lossless, dispersion-free channel, by directly reversing the time (time domain) or complex conjugating (frequency domain) of the received signal and retransmitting it, the wave will automatically focus back to the source position. The formula is:

[0137]

[0138] In the formula, For signals in the time domain, This is the signal after the time-domain signal is inverted. for The corresponding frequency domain signal, This is the signal obtained by taking the complex conjugate of the frequency domain signal. This is a Fourier transform.

[0139] In practical applications with dispersion, simple complex conjugation cannot achieve perfect focusing because the attenuation and dispersion during forward propagation are not compensated for. The signal undergoes... The channel effect is directly subjected to its complex conjugate. superscript The process involves taking the complex conjugate and transmitting it; its backpropagation will again experience channel effects. The final signal obtained is It is directly proportional, but contains serious distortion.

[0140] Complex cable networks introduce various distortion effects, including impedance discontinuities, frequency-varying losses, and frequency dispersion. An accurate model is needed to describe the electromagnetic characteristics of wave propagation in cable networks, providing a foundation for signal compensation and fault location. The cable network is abstracted as a weighted graph (nodes = endpoints, branches, terminations; edge weights = transmission matrix). Based on transmission line theory, the propagation function of each edge is calculated, and the overall transfer matrix is ​​assembled. During fault simulation, the established model is used to generate the forward transfer function. and reflection function .

[0141] Due to frequency-dependent attenuation and dispersion in cable links, directly retransmitting the received signal via complex conjugation often fails to achieve perfect focusing, necessitating compensation for channel distortion. Designing a suitable compensation function to correct attenuation and phase shift allows reverse propagation to simulate ideal channel conditions, thereby improving focusing quality.

[0142] In practical channels, the frequency-varying loss and dispersion introduced by forward propagation cause waveform distortion, and direct complex conjugate inversion leads to weakened focusing peaks and positional deviations. Therefore, a frequency domain compensation function is needed. multiplied by The inverted signal then approximately passes through an ideal inversion channel.

[0143] The parameterized compensation model includes amplitude compensation terms and phase compensation terms. The design in this application uses an exponential function:

[0144]

[0145] In the formula, For frequency domain compensation function, and This is the amplitude attenuation compensation parameter (can be positive or negative). and These are the phase delay compensation parameters. The total length of the theoretical propagation path, This is the set of compensation parameters.

[0146] The goal of the compensation function is to approximate the transmission function of an ideal channel as closely as possible, so that it can compensate for attenuation distortion and phase shift during forward propagation. The unit is Or dimensionless, Units are or (Depending on the frequency unit); the initial value is usually set to 0 (no compensation), and the optimal value is obtained through optimization. The initial value of the parameters can be selected according to the specific cable material and frequency band.

[0147] In the frequency domain inversion step, the received signal First, take the complex conjugate to obtain the received signal after complex conjugation. Then multiply by the compensation function The corrected inversion signal is obtained. In the time domain, this is equivalent to filtering out attenuation and adding phase delay. Parameter set The average value can be set by laboratory simulations, and then iteratively optimized by the DBO algorithm to obtain the final value. :

[0148]

[0149] In the formula, The optimized inversion signal, To obtain the complex conjugate received signal.

[0150] The design goal of the compensation function is to approximate the transfer function of the ideal time-inverted channel as closely as possible, and it is set as follows:

[0151]

[0152] In the formula, This represents the phase shift of the signal during propagation.

[0153] The inverse signal in the time domain is obtained through inverse Fourier transform. :

[0154]

[0155] In the formula, For Fourier transform, superscript This is the inverse operation. For time.

[0156] The goal of optimization is to make the inversion signal At the time location corresponding to the fault point This produces a peak that is as sharp as possible.

[0157] The high PSR candidate peak value is:

[0158]

[0159] In the formula, The peak-to-side-lobe ratio, The main valve window, based on the theoretical failure time A very narrow time window centered on The sidelobe region is the entire area on the time axis excluding the main lobe window, or the area after deliberately excluding the main lobe and the noise floor.

[0160] In one embodiment of this application, such as Figure 4 As shown, the DBO algorithm is used for parameter optimization. Several parallel behaviors are employed for global and local searches, with rolling ball behavior, dancing behavior, breeding behavior, and stealing behavior simulating different optimization strategies to improve solution quality. Finally, the optimal solution is found through iterative optimization.

[0161] The purpose of DBO is to find a set of parameters. To maximize PSR, the objective function is defined as follows:

[0162]

[0163] Through iterative optimization, the parameter combination that maximizes PSR is found.

[0164] Specifically:

[0165] A1: Randomly generated The initial position (parameter set) of each dung beetle (candidate solution). ), and calculate the objective function value corresponding to each candidate solution position. .

[0166] A2: Rolling ball behavior (global exploration):

[0167] The dung beetle updates its position with random step size and direction based on the current optimal solution and the global optimal solution. Simulating the process of a dung beetle rolling a dung ball to find new food sources, a wide-area search is implemented, and the update formula is:

[0168]

[0169] In the formula, For the first The candidate solution at the th... The set of compensation parameters for the next iteration For the first The candidate solution at the th... The set of compensation parameters for the next iteration This is the globally optimal solution. This is the current optimal solution. and This is the step size control factor. It is a random vector. This indicates element-wise multiplication.

[0170] A3: Dancing behavior (partial):

[0171] To avoid getting trapped in local optima, dung beetles "dance" with a certain probability, that is, they make small-range random perturbations around their current position, simulating the process of a dung beetle repositioning itself when encountering obstacles, performing a local fine search, and the update formula is:

[0172]

[0173] In the formula, The disturbance intensity factor, A standard normally distributed random vector, It is an identity matrix.

[0174] A4: Reproductive Behavior (Elite Preservation and Exploration):

[0175] Retain the current best dung beetle and its surrounding area, generate new dung beetles using Gaussian perturbation, simulate reproduction to ensure the inheritance and preservation of superior individuals, and conduct small-scale exploration. The update formula is:

[0176]

[0177] In the formula, This is the updated set of compensation parameters after Gaussian perturbation. is the Gaussian perturbation scaling factor.

[0178] A5: Theft (from local to global):

[0179] Simulating the behavior of a small dung beetle stealing dung balls from a large dung beetle, a new random exploration direction is introduced to help the algorithm escape local optima. The update formula is as follows:

[0180]

[0181] In the formula, and Step size factor For a uniformly distributed random vector, The position of another randomly selected candidate solution.

[0182] A6: Iteration: Repeat A2-A5 until the maximum number of iterations is reached or the convergence condition is met. This yields the optimal parameter vector, where, It is a very small positive number.

[0183] S5: Based on the optimized inversion signal, obtain high PSR candidate peak values, and backtrack to the original received signal to calculate the local energy within the first traveling wave theoretical time window of the candidate fault point. Select the candidate point with the largest energy as the real fault point.

[0184] In one embodiment of this application, even if a high PSR focused peak is obtained through optimization, spurious peaks may still be generated by noise or multiple reflections. To further verify the correctness of the peak values, a local energy extraction mechanism is introduced: for each high PSR candidate peak, its local energy in the original received signal is calculated to confirm the true fault location.

[0185] False peaks typically lack a local energy distribution that corresponds to the main waveform. By retracing back to the time domain of the original signal and extracting the energy of the first traveling wave corresponding to the candidate peak, the reliability of the judgment can be enhanced.

[0186] For each candidate peak time (Corresponding to potential fault location) ), calculate the energy of the candidate peak in the inverted signal, and backtrack to the original received signal to calculate the theoretical time window of the first traveling wave at the candidate fault point. Local energy within :

[0187]

[0188] In the formula, The original received signal, This is a tiny time interval used for local energy calculations within a signal time window, ensuring that only the first wavefront is captured.

[0189] Final fault determination criterion: Among all candidate peak values ​​where PSR exceeds the threshold, the maximum local energy of the corresponding initial traveling wave is selected as the true fault point.

[0190]

[0191] In the formula, This represents the actual point of failure.

[0192] After obtaining the final set of compensation parameters and injecting the inversion signal, several candidate peaks are identified (based on PSR threshold or peak amplitude). For each candidate, the corresponding window is found in the original received waveform according to the theoretical traveling wave delay, and the local energy is calculated. Finally, the candidate with the highest energy is determined as the real fault point, effectively reducing the risk of misjudgment caused by noise and multiple reflections.

[0193] This application provides a fault location method for EMTR cables based on the dung beetle optimization algorithm. The dung beetle optimization (DBO) algorithm is introduced into EMTR cable fault location: by simulating the behaviors of dung beetles—rolling a ball (global exploration), dancing (local fine-grained search), breeding (elite preservation and small-scale exploration), and stealing (escaping local optima)—automatic optimization of key EMTR compensation parameters is achieved, replacing traditional manual experience and solving the problems of low efficiency and poor robustness in parameter adjustment under complex scenarios. Furthermore, a parameterized compensation model for frequency-varying losses is designed: based on the frequency domain complex conjugate operation of electromagnetic time inversion, a compensation function including amplitude attenuation compensation, phase delay compensation, and nonlinear compensation terms is introduced. This can accurately offset attenuation distortion and phase shift during signal forward propagation, correcting the deficiency of traditional EMTR "simple complex conjugate inversion" in dealing with actual channel distortion. Meanwhile, the Peak-to-Side-Lobe Ratio (PSR) is used as the optimization objective function of the DBO algorithm: the energy ratio of the main lobe window to the side lobe region is taken as PSR. By using the DBO algorithm to maxPSR, the sharpness of the inverted signal at the fault point is enhanced, effectively suppressing noise and spurious peak interference caused by multiple reflections, thus solving the problem of easy misjudgment when taking the maximum peak in the traditional EMTR algorithm. In addition, a local energy extraction mechanism is constructed. The high PSR candidate peaks obtained after DBO optimization are traced back to the original received signal to calculate the local energy within the first traveling wave theoretical time window of the candidate fault point. Finally, the candidate point with the largest energy is selected as the real fault point, further reducing the risk of spurious peak misjudgment.

[0194] It should be noted that those skilled in the art will recognize that the embodiments described herein are for the purpose of helping readers understand the principles of this application, and should be understood as not limiting the scope of protection of this application to such specific statements and embodiments. Those skilled in the art can make various other specific modifications and combinations based on the technical teachings disclosed in this application without departing from the essence of this application, and these modifications and combinations are still within the scope of protection of this application.

Claims

1. A fault location method for EMTR cables based on a dung beetle optimization algorithm, characterized in that, include: S1: Construct a cable network topology model for the cable network; S2: Inject the detection signal into the cable network topology model, propagate it from the test end of the cable network topology model to the fault point, and then bounce back to the test end to receive the reflected signal containing fault characteristic information; S3: Perform time-reversal processing on the reflected signal, construct a frequency domain compensation function, use the frequency domain compensation function to correct the distortion of the inverted signal caused by attenuation and dispersion in the actual signal channel, and introduce the dung beetle optimization algorithm to iteratively optimize the compensation parameters of the frequency domain compensation function to obtain the optimized inverted signal; S4: Based on the optimized inversion signal, high PSR candidate peak values ​​are obtained, and the original received signal is traced back to calculate the local energy within the first traveling wave theoretical time window of the candidate fault point. The candidate point with the largest energy is selected as the real fault point. The process involves performing time-reversal processing on the reflected signal, constructing a frequency domain compensation function, and using this function to correct the distortion of the inverted signal caused by attenuation and dispersion in the actual signal channel. A dung beetle optimization algorithm is then introduced to iteratively optimize the compensation parameters of the frequency domain compensation function, resulting in an optimized inverted signal, including: S301: Perform time-reversal processing on the reflected signal and retransmit it; S302: Construct a frequency domain compensation function that includes amplitude compensation and phase compensation terms: In the formula, For frequency domain compensation function, Angular frequency, The imaginary unit, and For amplitude attenuation compensation parameters, and These are the phase delay compensation parameters. The total length of the theoretical propagation path, For the set of compensation parameters; S303: In the frequency domain, the received inversion signal is taken as its complex conjugate and multiplied by the compensation function to obtain the corrected inversion signal. Then, for the set of compensation parameters, the compensation parameters of the frequency domain compensation function are iteratively optimized using the dung beetle optimization algorithm to obtain the optimized inversion signal. : In the formula, The inversion signal obtained by taking the complex conjugate is... , For the test end propagation to the fault point The forward transfer function, Fault point Reflection propagates to the test end The reverse transfer function, This refers to the phase shift of the signal during propagation; S304: Based on The inverse signal in the time domain is obtained using the inverse Fourier transform. : In the formula, For Fourier transform, superscript This is the inverse operation. For time.

2. The EMTR cable fault location method based on the dung beetle optimization algorithm according to claim 1, characterized in that, The construction of a cable network topology model for the cable network includes: S101: Abstracting the cable network into a weighted graph model : In the formula, For a set of nodes, For the test end, As a branch point, The fault point For terminal nodes, For edge set, For nodes To the node cable section, For cable section In frequency The edge weight function below; S102: Model each segment of the uniform transmission cable as a uniform transmission line, and construct the voltage and current fluctuation equations for the cable line based on the telegraph equations: In the formula, It is a voltage wave. It is a current wave. Angular frequency, For location, The imaginary unit, , , and These are resistance, inductance, conductance, and capacitance per unit length, respectively. Solving the voltage and current fluctuation equations of the cable, we obtain the expression for the voltage wave: In the formula, Forward traveling wave, It is a reverse traveling wave. The propagation constant; Solve for the propagation constant based on the expression for voltage waves. and characteristic impedance : In the formula, The attenuation constant is It is the phase constant; S103: Treat each cable segment as a two-port network and define the scattering parameters. Characterizing the discontinuity in characteristic impedance, which leads to wave reflection and transmission, for Port network, scattering parameters The expression is: In the formula, For port The normalized incident wave, For port The normalized reflected wave, Indicates from port Voltage waves incident on the network, Indicates from port Voltage waves reflected back into the network For port Characteristic impedance, Indicates from port Incident wave to port Scattering parameters of the emitted wave; The scattering parameters of the two-port network are: In the formula, For series impedance, Characteristic impedance; Among them, port Reflectance coefficient for: In the formula, Indicates from port Incident wave pairs with port Scattering parameters of the emitted wave.

3. The EMTR cable fault location method based on the dung beetle optimization algorithm according to claim 2, characterized in that, The frequency domain expression of the reflected signal is: In the formula, The frequency domain representation of the reflected signal. For the frequency domain representation of the injected signal, For the test end propagation to the fault point The forward transfer function, This is the corrected reflection coefficient. and The key parameter for the reflection coefficient, Fault point Reflection propagates to the test end The reverse transfer function, To correct for the effects of noise and stray reflections; The total propagation delay is: In the formula, For the signal at group velocity Along path length The time required for dissemination This is the average operator. For path.

4. The EMTR cable fault location method based on the dung beetle optimization algorithm according to claim 3, characterized in that, The method of iteratively optimizing the compensation parameters of the frequency domain compensation function using the dung beetle optimization algorithm includes: A1: Initialize the parameters of the dung beetle optimization algorithm and generate them randomly. The initial positions of the candidate solutions are calculated, and the objective function value corresponding to each candidate solution position is calculated. ; A2: Update the position based on the current optimal solution and the global optimal solution, with a random step size and direction. The update formula is as follows: In the formula, For the first The candidate solution at the th... The set of compensation parameters for the next iteration For the first The candidate solution at the th... The set of compensation parameters for the next iteration This is the globally optimal solution. This is the current optimal solution. and This is the step size control factor. It is a random vector. This represents element-wise multiplication; A3: To avoid getting trapped in local optima, a small-scale random perturbation is made around the current position to simulate the process of a dung beetle repositioning itself when encountering an obstacle, performing a fine-grained local search. The update formula is: In the formula, The disturbance intensity factor, A standard normally distributed random vector, It is the identity matrix; A4: Retain the current optimal candidate solution and its surrounding region, generate new candidate solutions through Gaussian perturbation, simulate dung beetle reproduction, ensure the inheritance and preservation of excellent individuals, and conduct small-scale exploration. The update formula is: In the formula, This is the updated set of compensation parameters after Gaussian perturbation. The Gaussian perturbation scaling factor; A5: Simulating the behavior of a small dung beetle stealing dung balls from a large dung beetle, a new random exploration direction is introduced to escape local optima. The update formula is: In the formula, and Step size factor It is a uniformly distributed random vector; A6: Repeat A2-A5 until the maximum number of iterations is reached or the convergence condition is met. This yields the optimal parameter vector, where, It is a very small positive number.

5. The EMTR cable fault location method based on the dung beetle optimization algorithm according to claim 4, characterized in that, Based on the optimized inversion signal, high PSR candidate peak values ​​are obtained, and the original received signal is traced back to calculate the local energy within the first traveling wave theoretical time window of the candidate fault point. The candidate point with the highest energy is selected as the real fault point, including: S401: Inversion signal in the time domain obtained using inverse Fourier transform High PSR candidate peak values ​​were obtained: In the formula, The peak-to-side-lobe ratio, Main petal window This is the side lobe region; S402: Time for each candidate peak Calculate the energy of the candidate peak in the inverted signal, and backtrack to the original received signal to calculate the theoretical time window of the first traveling wave at the candidate fault point. Local energy within : In the formula, The original received signal, It is an extremely small time interval; S403: Among all candidate peak values ​​where PSR exceeds the threshold, select the maximum local energy of the corresponding initial traveling wave as the true fault point: In the formula, This represents the actual point of failure.

Images