Partial discharge signal detection and positioning method for power distribution cable intermediate joint

Abstract

The application discloses a partial discharge signal detection and positioning method for power distribution cable intermediate joints and relates to the technical field of power distribution equipment state monitoring, which comprises the following steps: collecting signal propagation characteristics of a partial discharge signal through a reference point array arranged around a power distribution cable intermediate joint; constructing a discharge source spatial positioning equation set and solving the discharge source spatial positioning equation set to obtain an initial solution; analyzing time-frequency characteristics of the partial discharge signal, identifying and classifying the partial discharge signal, and obtaining a discharge type judgment result; determining corresponding position correction parameters based on the discharge type judgment result; and fusing the position correction parameters and the initial solution, continuously correcting a positioning result through a recursive iteration optimization algorithm until the positioning error converges to a set threshold. The application realizes accurate mapping of discharge types and position characteristics, reduces the positioning error, improves the adaptability and convergence of the algorithm, and enhances the anti-interference capability in a complex electromagnetic environment.

Description

Technical Field

[0001] This invention relates to the field of power distribution equipment condition monitoring technology, specifically a method for detecting and locating partial discharge signals at intermediate joints of power distribution cables. Background Technology

[0002] Partial discharge detection and location technology is an important means of assessing the insulation condition of cable joints. Traditional partial discharge detection mainly relies on pulsed current and ultrasonic methods, which are relatively mature in laboratory environments. However, with the intelligent development of power distribution networks, higher requirements are placed on the accuracy, real-time performance, and adaptability of partial discharge detection and location. Especially in areas with complex structures and multiple layers of insulation, such as cable joints, traditional detection methods have significant limitations in terms of location accuracy and discharge type identification. Ultrasonic methods are greatly affected by reflections from dielectric boundaries, resulting in a significant decrease in location accuracy in multi-layered insulation structures; while pulsed current methods, although highly sensitive, struggle to accurately distinguish different types of discharge sources and are easily affected by external interference in complex electromagnetic environments.

[0003] Currently, the commonly used positioning methods in the industry are mainly based on time difference and energy attenuation characteristics. However, these methods often treat different types of discharge sources as equal, ignoring the differences in spatial distribution and signal characteristics among various discharges. Furthermore, existing technologies lack an effective mapping mechanism between signal feature extraction and spatial positioning calculations, leading to deviations between the positioning results and the actual physical characteristics of the discharge. Particularly when dealing with different types of discharges such as internal discharges, surface discharges, levitation discharges, and corona discharges, the positioning accuracy and reliability fail to meet engineering requirements due to the failure to fully consider their unique spatial distribution patterns and signal propagation characteristics. Summary of the Invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the present invention provides a method for detecting and locating partial discharge signals at intermediate joints of power distribution cables, which can solve the problems mentioned in the background art.

[0006] To address the aforementioned technical problems, this invention provides the following technical solution: a method for detecting and locating partial discharge signals at intermediate joints of power distribution cables, comprising: acquiring signal propagation characteristics of partial discharge signals by means of a reference point array set around the intermediate joint of the power distribution cable; constructing a set of spatial positioning equations for the discharge source based on the signal propagation characteristics, and solving the set of spatial positioning equations to obtain an initial solution; analyzing the time-frequency characteristics of the partial discharge signals, and identifying and classifying the partial discharge signals to obtain a discharge type determination result; determining corresponding position correction parameters based on the discharge type determination result; fusing the position correction parameters with the initial solution, and continuously correcting the positioning result through a recursive iterative optimization algorithm until the positioning error converges to within a set threshold.

[0007] As a preferred embodiment of the method for detecting and locating partial discharge signals at intermediate joints of power distribution cables according to the present invention, the reference point array includes multiple electrode sensors arranged in a ring, and each electrode sensor amplifies and filters the acquired partial discharge signal through a signal conditioning circuit; the signal propagation characteristics include propagation time difference and energy attenuation characteristics.

[0008] As a preferred embodiment of the method for detecting and locating partial discharge signals at intermediate joints of power distribution cables according to the present invention, the method includes the following steps: Based on the signal propagation characteristics, a set of spatial location equations for the discharge source is constructed, and the set of spatial location equations for the power source is solved to obtain an initial solution.

[0009] Establish a set of positioning equations based on the propagation time difference. Specifically, assume the coordinates of the discharge source are (x0, y0, z0) and the coordinates of the i-th sensor are (xᵢ, yᵢ, zᵢ). Then the signal propagation time difference satisfies:

[0010] ;

[0011] in, Let v be the time difference between the arrival of the signal at the i-th sensor and the j-th sensor, and let v be the propagation speed of the electromagnetic wave in the medium. , ∈[1,N] and i≠j, where N is the total number of sensors;

[0012] The spatial positioning equations of the discharge source are expressed as follows:

[0013] ;

[0014] in, , , are the energy values ​​of the signal when it reaches the i-th and j-th sensors, respectively, and α is the attenuation coefficient of the medium. , These are the distances from the power source to the i-th and j-th sensors, respectively.

[0015] Construct a composite error function:

[0016] ;

[0017] Where m represents the measured value and c represents the calculated value. and These are the weighting coefficients. As a balance factor;

[0018] The initial solution is obtained by iterative least squares method; the initial solution represents the initial spatial coordinates of the discharge source.

[0019] As a preferred embodiment of the method for detecting and locating partial discharge signals at intermediate joints of power distribution cables according to the present invention, the initial solution obtained by the iterative least squares method is expressed by the following formula:

[0020] ;

[0021] in, Indicates the number of iterations. For Jacobian matrices, This is the weight matrix. For the residual vector, Step size factor;

[0022] Order No. The weighted sum of squared residuals for the next iteration is :

[0023] ;

[0024] If the weighted sum of squared residuals in the current iteration Less than the average of the previous two iterations, that is:

[0025] ;

[0026] Then let the step size factor Represented as:

[0027] ;

[0028] Otherwise, let the step size factor be... Represented as:

[0029] .

[0030] As a preferred embodiment of the method for detecting and locating partial discharge signals at intermediate joints of power distribution cables according to the present invention, the method includes the following steps: analyzing the time-frequency characteristics of the partial discharge signal, identifying and classifying the partial discharge signal to obtain a discharge type determination result.

[0031] Wavelet transform is used to decompose the partial discharge signal into multiple scales and extract time-frequency features. The partial discharge signal is decomposed into high-frequency band, mid-frequency band and low-frequency band, and the detail coefficients and approximation coefficients of each frequency band are obtained. Based on the detail coefficients and approximation coefficients, the energy distribution characteristics of the partial discharge signal are obtained.

[0032] A time-frequency feature vector is constructed based on the energy distribution characteristics. The time-frequency feature vector includes the energy ratio of the high-frequency band to the low-frequency band, the pulse feature function, and the waveform statistical characteristics. The energy ratio of the high-frequency band to the low-frequency band is determined by wavelet coefficients and energy attenuation characteristics. The pulse feature function is composed of amplitude distribution and time characteristics. The waveform statistical characteristics include waveform skewness and waveform kurtosis.

[0033] Discharge type identification is performed based on the aforementioned time-frequency feature vector, with the following criteria:

[0034] If the energy ratio of the high frequency band to the low frequency band Greater than the unit value, and the pulse characteristic function If the global maximum value is obtained within the positive half-cycle of the power frequency, and the waveform skewness value is positive, then it is determined to be internal discharge.

[0035] If the energy ratio of the high frequency band to the low frequency band Less than the unit value, and the pulse characteristic function If the global maximum value is obtained within the negative half-cycle of the power frequency, and the waveform skewness value is negative, then it is determined to be surface discharge.

[0036] If the pulse characteristic function If multiple peaks occur within one power frequency cycle, and the amplitude ratio of adjacent peaks deviates from the unit value to a degree within a first preset range, and the phase difference of the peaks is greater than the basic segmentation unit of the power frequency cycle, then it is determined to be a floating discharge.

[0037] If the pulse characteristic function If multiple consecutive peaks are observed within a local phase interval, and the peak sequence follows an exponential decay law, while the phase difference between adjacent peaks is less than the minimum segmentation unit of the power frequency cycle, then it is determined to be corona discharge.

[0038] Based on the above criteria, the discharge type determination result is obtained.

[0039] As a preferred embodiment of the method for detecting and locating partial discharge signals at intermediate joints of power distribution cables according to the present invention, the method for determining corresponding position correction parameters based on the discharge type determination result includes the following steps:

[0040] A discharge location feature model is constructed; the discharge location feature model includes a location feature parameter matrix for different discharge types, and the location feature parameter matrix is ​​established based on the high-low frequency band energy ratio, pulse feature function, and waveform statistical features.

[0041] Establish coordinate transformation relationships between the rectangular coordinate system and the cylindrical coordinate system;

[0042] Constructing basic location feature functions :

[0043] ;

[0044] in, For location feature functions, This is the normalized value of the energy ratio between high and low frequency bands. The normalized impulse characteristic function, For waveform skewness, For waveform kurtosis, , , The feature weight index;

[0045] If the discharge type determination result is internal discharge or surface discharge, then a radially dominant characteristic function is constructed. :

[0046] ;

[0047] in, Radial attenuation coefficient; It is a radial correction factor;

[0048] If the discharge type determination result is a floating discharge type, then a balanced characteristic function is constructed. :

[0049] ;

[0050] in, For radial feature scale, Angular feature scale, As an axial characteristic scale, As a balance factor;

[0051] If the discharge type determination result is corona discharge, then an axial dominant characteristic function is constructed. :

[0052] ;

[0053] in, Angular modulation coefficient; The modulation coefficient is the height. This is the axial correction factor;

[0054] Extract the correction parameter matrix and transform it from cylindrical coordinates to Cartesian coordinates;

[0055] The position correction parameters are calculated using matrix multiplication based on the correction parameter matrix in the rectangular coordinate system and the spatial coordinate values ​​to be corrected.

[0056] As a preferred embodiment of the method for detecting and locating partial discharge signals at intermediate joints of power distribution cables according to the present invention, the method involves fusing the position correction parameters with the initial solution and continuously correcting the positioning result through a recursive iterative optimization algorithm until the positioning error converges to a set threshold. This includes the following steps:

[0057] The position correction parameters and the initial solution are superimposed and fused in a weighted manner to obtain the initial value for iteration;

[0058] Construct an error evaluation function for iterative optimization;

[0059] The iteration parameters are updated based on the calculation results of the error evaluation function. The iteration stops when the relative rate of change of the error evaluation function is less than a preset threshold. Otherwise, the weight coefficient is increased or decreased by comparing the error between two adjacent iterations until the final positioning result is obtained.

[0060] To further address the aforementioned technical problems, this invention provides the following technical solution: a partial discharge signal detection and positioning system for intermediate joints of power distribution cables, comprising: a signal acquisition module for acquiring signal propagation characteristics of partial discharge signals through a reference point array set around the intermediate joints of power distribution cables; a spatial positioning module for constructing a set of spatial positioning equations for the discharge source based on the signal propagation characteristics, and solving the set of spatial positioning equations to obtain an initial solution; a pattern recognition module for analyzing the time-frequency characteristics of the partial discharge signals, and identifying and classifying the partial discharge signals to obtain a discharge type determination result; a parameter extraction module for determining corresponding position correction parameters based on the discharge type determination result; and an optimization iteration module for fusing the position correction parameters with the initial solution, and continuously correcting the positioning result through a recursive iterative optimization algorithm until the positioning error converges to within a set threshold.

[0061] A computer device includes a memory and a processor, the memory storing a computer program, characterized in that the processor executes the computer program to implement the steps of the method for detecting and locating partial discharge signals of intermediate joints in power distribution cables as described above.

[0062] A computer-readable storage medium having a computer program stored thereon, characterized in that, when the computer program is executed by a processor, it implements the steps of the method for detecting and locating partial discharge signals at intermediate joints of power distribution cables as described above.

[0063] The beneficial effects of this invention are as follows: This invention employs a ring-distributed reference point array in conjunction with a dedicated signal conditioning circuit, overcoming the signal distortion problem inherent in traditional acquisition methods under complex cable joint structures. By combining propagation time difference and energy attenuation characteristics to construct a composite error function, the reliability of the initial solution is improved. This invention proposes a family of characteristic functions based on discharge type, effectively addressing the specificity issues of different types of discharge sources in spatial positioning by constructing radially dominant, balanced, and axially dominant characteristic functions. Finally, this invention achieves deep fusion of characteristic quantities and spatial gradients through the design of a multi-level correction parameter matrix, overcoming the technical bottleneck of the separation of characteristic information and spatial information in traditional methods. Attached Figure Description

[0064] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0065] Figure 1 This is a schematic diagram of the overall process for the detection and location of partial discharge signals at intermediate joints of power distribution cables proposed in this invention.

[0066] Figure 2 This is an interactive schematic diagram of the partial discharge signal detection and location system for intermediate joints of power distribution cables proposed in this invention.

[0067] Figure 3 This is a diagram of the computer equipment used in the method for detecting and locating partial discharge signals at intermediate joints of power distribution cables proposed in this invention. Detailed Implementation

[0068] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0069] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0070] Example 1, referring to Figure 1 As an embodiment of the present invention, a method for detecting and locating partial discharge signals at intermediate joints of power distribution cables is provided.

[0071] Figure 1 A schematic diagram of the overall process for detecting and locating partial discharge signals at intermediate joints of power distribution cables is shown, including the following steps:

[0072] S1: The signal propagation characteristics of partial discharge signals are collected by an array of reference points set around the intermediate joints of the power distribution cable.

[0073] It should be noted that the reference point array includes multiple electrode sensors arranged in a ring. Each electrode sensor amplifies and filters the acquired partial discharge signal through a signal conditioning circuit. The signal propagation characteristics include propagation time difference and energy attenuation characteristics.

[0074] Furthermore, a complete signal acquisition system needs to be constructed first. To this end, multiple electrode sensors are arranged in a reference point array structure at the circumferential spatial location of the power distribution cable joint. The electrode sensors have a ring structure, allowing them to uniformly cover the outer surface of the cable joint, ensuring omnidirectional acquisition of partial discharge signals. This ring-shaped spatial arrangement enables synchronous monitoring of partial discharge signals at different spatial locations, providing fundamental data support for subsequent signal localization calculations.

[0075] Each electrode sensor in the reference point array is equipped with a dedicated signal conditioning circuit for processing the acquired raw partial discharge signal. The signal conditioning circuit includes a preamplifier and a bandpass filter. The preamplifier enhances the amplitude of the weak discharge signal, while the bandpass filter suppresses power frequency interference and high-frequency noise, extracting the effective frequency band components of the partial discharge signal. After signal conditioning, the electrode sensors can accurately acquire the propagation time difference and energy attenuation characteristics of the partial discharge signal. These two characteristics contain information about the spatial location of the discharge source. Specifically, the propagation time difference reflects the timing of the discharge signal's arrival at each sensor, while the energy attenuation characteristic reflects the intensity variation of the signal during its spatial propagation.

[0076] S2: Based on the signal propagation characteristics, a set of spatial positioning equations for the discharge source is constructed, and the initial solution of the set of spatial positioning equations for the discharge source is solved by iterative least squares method.

[0077] Specifically, the initial solution represents the initial spatial coordinates of the power source.

[0078] S2.1: Establish a set of positioning equations based on the propagation time difference.

[0079] Assuming the coordinates of the power source are (x0, y0, z0) and the coordinates of the i-th sensor are (xᵢ, yᵢ, zᵢ), then the signal propagation time difference satisfies:

[0080] ;

[0081] in, Let v be the time difference between the arrival of the signal at the i-th sensor and the j-th sensor, and let v be the propagation speed of the electromagnetic wave in the medium. , ∈[1,N] and i≠j, where N is the total number of sensors.

[0082] S2.2: Establish the spatial positioning equations for the discharge source based on energy decay:

[0083] ;

[0084] in, , , are the energy values ​​of the signal when it reaches the i-th and j-th sensors, respectively, and α is the attenuation coefficient of the medium. , denoted as , and respectively as , the distances from the power source to the i-th and j-th sensors.

[0085] S2.3: Constructing the composite error function:

[0086] ;

[0087] Where m represents the measured value and c represents the calculated value. and These are the weighting coefficients. It is a balancing factor.

[0088] S2.4: Solve for the initial solution using the iterative least squares method:

[0089] ;

[0090] in, Indicates the number of iterations. For Jacobian matrices, This is the weight matrix. For the residual vector, This is the step size factor.

[0091] Order No. The weighted sum of squared residuals for the next iteration is :

[0092] ;

[0093] If the weighted sum of squared residuals in the current iteration is less than the average of the previous two iterations, that is:

[0094] ;

[0095] Then let:

[0096] ;

[0097] Otherwise, let:

[0098] .

[0099] In a specific embodiment of the present invention, the process of calculating the location of the discharge source based on the acquired signal propagation characteristics is as follows: First, a set of location equations based on time difference is established. Since the partial discharge signal experiences a time delay during propagation, different sensors receive the signal at different times; this time difference contains the location information of the discharge source. By analyzing the time difference between the arrival of the signal at adjacent sensors, constraint equations reflecting the spatial location of the discharge source can be constructed. Simultaneously, considering the energy attenuation characteristics of the discharge signal during propagation, a set of energy attenuation location equations is established. The signal energy attenuates with increasing propagation distance; this attenuation law also reflects the spatial relationship between the discharge source and each sensor. By analyzing the ratio of signal energy at different sensors, supplementary constraint conditions required for location calculation can be obtained. This dual-feature combination method has stronger anti-interference capabilities compared to traditional single-feature location methods.

[0100] In solving the positioning equations, this invention employs an iterative least squares method. First, a composite error function is constructed, organically fusing time difference and energy attenuation features. By appropriately setting weighting coefficients and balancing factors, the advantages of each feature in the positioning calculation are ensured. This feature fusion strategy avoids the positioning errors that may arise from relying on a single feature in traditional methods.

[0101] In the iterative solution process, this invention introduces an adaptive step size mechanism based on residual changes. The iteration step size is dynamically adjusted by comparing the weighted sum of squared residuals of the current iteration with those of the previous two iterations. When the residuals decrease rapidly, a larger step size is used to accelerate convergence; when the residual changes slowly, a smaller step size is used to improve accuracy. This adaptive mechanism effectively solves the problems of slow convergence or oscillations in traditional fixed-step-size algorithms.

[0102] Finally, through dual-feature constraints and adaptive iterative optimization, an initial solution reflecting the spatial location of the discharge source is obtained. This positioning method fully utilizes the complementarity of time difference and energy attenuation characteristics, and improves the accuracy of discharge source positioning in electromagnetic environments such as distribution cable intermediate joints through feature fusion and iterative optimization.

[0103] S3: Analyze the time-frequency characteristics of the partial discharge signal, identify and classify the partial discharge signal, and obtain the discharge type determination result.

[0104] S3.1: Wavelet transform is used to perform multi-scale decomposition of the partial discharge signal and extract time-frequency features. Wavelet basis functions with good local characteristics and symmetry are selected to decompose the signal into three frequency bands: high frequency, mid frequency, and low frequency. The detail coefficients and approximation coefficients of each frequency band are obtained to obtain the energy distribution characteristics of the signal.

[0105] Specifically, in the detection and localization of partial discharge signals at intermediate joints of power distribution cables, to effectively extract the time-frequency characteristics of the discharge signal, a multi-scale wavelet transform is first used to decompose the acquired partial discharge signal. Considering the non-stationary, transient, and multi-scale characteristics of the partial discharge signal, wavelet basis functions with good time-frequency localization properties, symmetry, and orthogonality are selected. These wavelet basis functions can effectively separate components of different frequency bands while maintaining the integrity of the signal characteristics.

[0106] In wavelet decomposition, the partial discharge signal is sequentially divided into three frequency bands: a high-frequency band (reflecting the rapid changes in the discharge pulse), a mid-frequency band (representing the main energy distribution of the discharge process), and a low-frequency band (characterizing the gradual trend of discharge development). By calculating the wavelet coefficients of each frequency band, detail coefficients and approximation coefficients of the signal at different scales are obtained. The detail coefficients reflect the local characteristics of the partial discharge signal in each frequency band, while the approximation coefficients characterize the overall trend of the signal. Based on these wavelet coefficients, the energy distribution patterns of the signal in different frequency bands are further analyzed, providing fundamental features for subsequent discharge type identification. This multi-scale decomposition method fully considers the multidimensional characteristics of partial discharge signals from distribution cable joints and can effectively extract the feature information required for discharge type determination.

[0107] S3.2: Construct time-frequency feature vectors.

[0108] a. Energy ratio of high-frequency band to low-frequency band:

[0109] ;

[0110] in, This represents the energy ratio between the high-frequency band and the low-frequency band. and These are the wavelet coefficients for the high-frequency band and the low-frequency band, respectively. The number of sampling points. , Let be the energy values ​​of the signal when it reaches the i-th and j-th sensors, respectively. , denoted as , and respectively as , the distances from the power source to the i-th and j-th sensors.

[0111] b. Pulse characteristics:

[0112] ;

[0113] in, For phase The pulse characteristic function at that location, Let be the amplitude of the i-th pulse. For the corresponding phase, For pulse width characteristic parameters, Let be the time difference between the arrival of the signal at the i-th sensor and the j-th sensor.

[0114] It should be noted that the pulse width characteristic parameter It describes the time duration characteristics of partial discharge pulses, reflecting the time scale of charge accumulation and release during the discharge process. Its value is determined by the physical characteristics of the discharge development process, and has different distribution patterns for different types of discharge mechanisms.

[0115] c. Waveform characteristics:

[0116] ;

[0117] in, For waveform skewness, For waveform kurtosis, For signal sample values, The mean, Here, M represents the standard deviation, and M represents the number of sampling points.

[0118] This invention combines the energy distribution characteristics obtained from wavelet decomposition with the energy attenuation characteristics during signal propagation. By introducing an energy attenuation compensation term, a mapping relationship between the energy ratio of high and low frequency bands and the discharge type is established. Simultaneously, the pulse characteristic function fully considers the time propagation characteristics, while the waveform statistical characteristics reflect the statistical regularities of different discharge types. This multi-dimensional feature fusion strategy overcomes the limitations of traditional single-feature methods.

[0119] S3.3: Discharge type identification is performed based on feature vectors, with the following criteria:

[0120] If the energy ratio of the high frequency band to the low frequency band Greater than the unit value, and the pulse characteristic function If the global maximum value is obtained within the positive half-cycle of the power frequency, and the waveform skewness value is positive, then it is determined to be internal discharge.

[0121] If the energy ratio of the high frequency band to the low frequency band Less than the unit value, and the pulse characteristic function If the global maximum value is obtained within the negative half-cycle of the power frequency, and the waveform skewness value is negative, then it is determined to be surface discharge.

[0122] If the pulse characteristic function If multiple peaks occur within one power frequency cycle, and the amplitude ratio of adjacent peaks deviates from the unit value by a degree within a first preset range, while the phase difference of the peaks is greater than the basic segmentation unit of the power frequency cycle, then it is determined to be a floating discharge; wherein, the first preset range can be plus or minus ten percent.

[0123] If the pulse characteristic function If multiple consecutive peaks are observed within a local phase interval, and the peak sequence follows an exponential decay law, while the phase difference between adjacent peaks is less than the minimum segmentation unit of the power frequency cycle, then it is determined to be corona discharge.

[0124] Based on the above criteria, the discharge type determination result is obtained.

[0125] It should be noted that the unit value refers to the standard reference value where the energy ratio or amplitude ratio is equal to 1. This value indicates that the two physical quantities being compared have equal magnitudes and serves as an important reference benchmark for determining different discharge mechanisms in discharge type identification.

[0126] Preferably, in the discharge type identification stage, this invention establishes a systematic determination rule based on the physical mechanism of discharge. By analyzing the typical characteristics of different types of discharge in the time-frequency domain, a complete determination criterion system is constructed. This invention avoids the subjectivity of traditional empirical thresholds and provides reliable discharge type identification results. In particular, the determination rule considers the combination relationship of multiple features, enhancing the robustness of the identification process.

[0127] S4: Based on the discharge type determination result, determine the corresponding position correction parameters.

[0128] S4.1: Construct a discharge location feature model, which includes location feature parameter matrices for different discharge types.

[0129] Among them, the location feature parameter matrix is ​​established based on the energy ratio of high and low frequency bands, pulse characteristic function and waveform statistical characteristics, reflecting the spatial distribution law of different types of discharge sources.

[0130] It should be noted that the discharge location feature model is a complete feature representation system, which includes the location feature parameter matrix as its core component. The location feature parameter matrix is ​​the specific data structure used for storage and calculation in the discharge location feature model.

[0131] S4.2: To ensure the uniformity of the coordinate system, coordinate transformation relationships are first established:

[0132] ;

[0133] in, , , These are the spatial coordinates in a rectangular coordinate system; , , These are the spatial position coordinates in a cylindrical coordinate system. Radial distance, It is the azimuth angle. The axial height is used. Compared to calculations performed directly in a Cartesian coordinate system, using a cylindrical coordinate system better reflects the geometric characteristics of cable joints and can more intuitively describe the spatial distribution of discharge sources.

[0134] Establish basic location feature functions :

[0135] ;

[0136] in, For location feature functions, It is the normalized value (between 0 and 1) of the energy ratio of high and low frequency bands. This is the normalized impulse characteristic function (between 0 and 1). For waveform skewness, For waveform kurtosis, , , The feature weight index is used to adjust the contribution ratio of each feature.

[0137] Preferably, this invention proposes a fundamental location feature function for multi-feature fusion. Compared with traditional methods, this function not only considers energy distribution characteristics but also introduces an exponential weighting mechanism for pulse features and waveform statistical features. Through dynamic adjustment of the three feature weight exponents, adaptive fusion of different feature quantities in positioning calculation is achieved. In practical applications, this multi-feature fusion strategy can effectively suppress positioning errors that may be caused by single features, improve the stability of positioning results, and the positioning accuracy after feature fusion is also improved compared to single feature fusion.

[0138] Specifically, the form of the fundamental position characteristic function differs depending on the discharge type. For internal and surface discharge types, a radially dominant characteristic function is constructed. :

[0139] ;

[0140] in, The radial attenuation coefficient describes the influence of axial height on radial distribution. It characterizes the energy loss of the discharge signal during radial propagation and ranges from 0.1 to 1.0. A larger value indicates slower radial propagation attenuation; The radial correction factor is used to correct the contribution weights at different radial positions. It is a piecewise function related to the radial distance r, and takes a larger value at the interface of the insulation layer.

[0141] Preferably, this invention improves the positioning accuracy of the internal discharge source by introducing an exponential radial attenuation term and a radial correction factor. Traditional methods often employ simple distance attenuation models, failing to fully consider the radial and axial coupling relationship of the internal discharge within the cable joint, resulting in significant positioning errors in the insulation delamination region. The radially dominant characteristic function proposed in this invention, through… The project establishes a secondary coupling relationship between axial height and radial distance, which is more consistent with the physical characteristics of electric field distribution. At the same time, a radial correction factor is introduced to dynamically weight the contribution of different radial positions, effectively solving the positioning deviation caused by the non-uniformity of the insulation structure.

[0142] For the suspended discharge type, construct an equalization characteristic function. :

[0143] ;

[0144] in, It is the radial characteristic dimension (usually 1 / 2 of the outer diameter of the joint). The angular feature scale (value 2π) is used. It is the axial characteristic dimension (usually 1 / 2 of the joint length). This is a balancing factor used to balance the component weights in the three directions, with a value ranging from 0.5 to 2.0. When the weight is 1, it means that the weights in the three directions are equal.

[0145] It should be noted that this invention designs an equilibrium characteristic function based on a normalized quadratic form, specifically targeting the spatial motion characteristics of levitated discharge. Compared with existing technologies, this function introduces characteristic scale parameters in three directions ( , , This approach achieves precise quantification of spatial anisotropy; simultaneously, it employs an equalization factor for overall modulation, overcoming the imbalance of contributions in different directions inherent in traditional methods. Furthermore, the selection of the quadratic structure fully considers the random ionization characteristics of suspended discharge within the cavity, avoiding directional deviations that may result from simple linear superposition.

[0146] For corona discharge types, construct axially dominant characteristic functions. :

[0147] ;

[0148] in, The angular modulation coefficient (ranging from 0.1 to 1) adjusts the degree of influence of azimuth angle changes. It describes the modulation depth of discharge intensity as the azimuth angle changes, and its value ranges from 0.1 to 1.0. The smaller the value, the more uniform the angular distribution. The high modulation coefficient (ranging from 1 to 10) controls the dominance of the axial component, ranging from 1.0 to 10.0. A larger value indicates stronger axial dominance. This is an axial correction factor used to correct the contribution weights at different height positions. It compensates for the non-uniformity of axial position and is a factor related to height. The piecewise continuous function takes a larger value at the joint end.

[0149] It should be noted that this invention, based on the physical mechanism of corona discharge, proposes a characteristic function with axial dominance. Traditional methods often employ uniform decay models, failing to reflect the gradual development characteristics of corona discharge along the axial direction. This invention, by designing a composite exponential term structure, unifies the radial and angular components under axial-scale modulation, maintaining the spatial directionality of corona discharge while considering the influence of radial diffusion. Furthermore, this invention achieves a non-uniform spatial distribution of discharge intensity through the combination of angular and height modulation coefficients; simultaneously, it introduces an axial correction factor to correct the axial position, improving the positioning accuracy of the boundary region.

[0150] S4.3: Extracting the correction parameter matrix based on location feature functions:

[0151] ;

[0152] in, It is a 3×3 correction parameter matrix. , , These are the partial derivatives of the characteristic functions, It is a 3×3 orthogonal transformation matrix from the feature space to the physical space, and each element represents the correction amount of the discharge characteristics to the spatial position.

[0153] Preferably, the multi-level correction parameter matrix structure provided by this invention, compared with traditional methods that only consider gradient information, establishes a deep fusion relationship between feature quantities and spatial gradients by introducing a row weighting mechanism of energy ratio and pulse characteristics. This ensures that the correction parameters not only include the changing trends of positional features but also incorporate the physical characteristic information of the discharge signal, improving the accuracy of the correction. Simultaneously, the introduction of an orthogonal transformation matrix achieves a unified mapping from feature space to physical space, ensuring the physical meaning of the correction parameters.

[0154] S4.4: Transform the correction parameter matrix from cylindrical coordinates to Cartesian coordinates:

[0155] ;

[0156] in, This is the correction parameter matrix in Cartesian coordinates, with the first row corresponding to... The direction correction amount, corresponding to the second row. The direction correction amount corresponds to the third row. The amount of correction for direction.

[0157] It should be noted that this invention employs a rotation matrix structure in coordinate transformation. Unlike traditional simple coordinate transformations, this matrix introduces dynamic compensation for radial distance on top of angular transformation. This design considers the nonlinear characteristics of the discharge source at different radial positions. Simultaneously, by maintaining... The independence of the directional components avoids the accumulation of errors during coordinate transformation.

[0158] S4.5: Based on the above calculation results, the position correction parameters in the spatial rectangular coordinate system are obtained:

[0159] ;

[0160] in, , , The final position correction parameters are denoted by , representing the correction amounts in the x, y, and z directions, respectively. , , These are the spatial coordinate values ​​to be corrected.

[0161] It should be noted that this invention uses matrix multiplication to achieve unified calculation of position correction parameters. This method is not only simple and intuitive in form, but more importantly, it achieves coupled calculation of corrections in each direction through matrix operations, overcoming the cumulative errors that may be caused by independent corrections in each direction in traditional methods.

[0162] Ideally, step S4 effectively addresses the technical problems of single features and poor adaptability in traditional positioning methods. The exponential weighted fusion mechanism introduced into the basic position feature function enables adaptive combination of different feature quantities, significantly improving the reliability of the positioning results. The feature function design for different discharge types fully considers the physical mechanism, giving the positioning results clear physical meaning. The correction parameter matrix innovatively achieves deep fusion of feature quantities and spatial gradients, solving the problem of the separation between feature information and spatial information. At the same time, the improved coordinate transformation method ensures the spatial consistency of the positioning results.

[0163] S5: The position correction parameters are fused with the initial solution, and the positioning result is continuously corrected through a recursive iterative optimization algorithm until the positioning error converges to within the set threshold.

[0164] S5.1: The position correction parameters and the initial solution are superimposed and fused in a weighted manner to obtain the initial value for iteration.

[0165] Two weighting coefficients are used to adjust the contribution ratio of the position correction parameter and the historical iteration amount, respectively. The initial value of the weighting coefficient can be preset based on experience.

[0166] S5.2: Construct an iteratively optimized error evaluation function.

[0167] The error evaluation function includes a mean squared error term and a norm constraint term. The mean squared error term is used to evaluate positioning accuracy, while the norm constraint term is used to prevent iterative divergence. The coefficients of the constraint term can be preset based on experience.

[0168] S5.3: Update the iteration parameters based on the calculation results of the error evaluation function. Stop the iteration when the relative rate of change of the error evaluation function is less than the preset threshold. Otherwise, increase or decrease the value of the weight coefficient by comparing the error of two adjacent iterations until the final positioning result is obtained.

[0169] It should be noted that in the partial discharge localization process of distribution cable intermediate joints, the preset threshold is a key parameter for evaluating the convergence of iterative calculations. This threshold reflects the error tolerance between two adjacent iterations, and its value needs to be balanced between positioning accuracy and computational efficiency. Based on the structural characteristics of cable intermediate joints and actual engineering requirements, the preset threshold is usually set in the form of a relative error. Specifically, when the relative rate of change of the error evaluation function is less than this threshold, the iterative calculation is considered to have reached convergence. Considering the typical dimensions of distribution cable intermediate joints (diameter 50-150mm, length 300-500mm) and the accuracy requirements for partial discharge localization (usually requiring the positioning error to not exceed 5% of the joint size), and based on experimental verification, this threshold is preferably set within the range of 0.5%-2%. An excessively large threshold will reduce positioning accuracy, while an excessively small threshold will lead to a significant increase in computational load and may result in numerical oscillations.

[0170] In summary, this invention employs a ring-shaped array of reference points in conjunction with a dedicated signal conditioning circuit, overcoming the signal distortion problem inherent in traditional acquisition methods under complex cable joint structures. By combining propagation time difference and energy attenuation characteristics to construct a composite error function, the reliability of the initial solution is improved. This invention proposes a family of characteristic functions based on discharge type, effectively addressing the specificity issues of different discharge types in spatial positioning by constructing radially dominant, balanced, and axially dominant characteristic functions. Finally, this invention achieves deep fusion of characteristic quantities and spatial gradients through the design of a multi-level correction parameter matrix, overcoming the technical bottleneck of the separation of characteristic information and spatial information in traditional methods.

[0171] Example 2, an embodiment of the present invention, provides a system for detecting and locating partial discharge signals at intermediate joints of power distribution cables, such as... Figure 2 The diagram shown is an interactive illustration of the system, including:

[0172] The signal acquisition module is used to acquire the signal propagation characteristics of partial discharge signals by means of an array of reference points set around the intermediate joint of the power distribution cable;

[0173] The spatial positioning module is used to construct a set of spatial positioning equations for the power source based on signal propagation characteristics, and to solve the set of spatial positioning equations for the power source to obtain the initial solution.

[0174] The pattern recognition module is used to analyze the time-frequency characteristics of partial discharge signals, identify and classify partial discharge signals, and obtain discharge type determination results.

[0175] The parameter extraction module determines the corresponding position correction parameters based on the discharge type determination result;

[0176] The optimization iteration module is used to fuse the position correction parameters with the initial solution and continuously correct the positioning result through a recursive iterative optimization algorithm until the positioning error converges to within a set threshold.

[0177] Example 3, referring to Figure 3 This is one embodiment of the present invention, which differs from the previous embodiment in that: if the function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, essentially, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes: USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, optical disks, and other media capable of storing program code.

[0178] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-including system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0179] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0180] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0181] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for detecting and locating partial discharge signals in power cable joint, characterized in that, include: The signal propagation characteristics of partial discharge signals are collected by setting up an array of reference points around the intermediate joints of power distribution cables; Based on the signal propagation characteristics, a set of spatial positioning equations for the power source is constructed, and the set of spatial positioning equations for the power source is solved to obtain an initial solution; The time-frequency characteristics of the partial discharge signal are analyzed, and the partial discharge signal is identified and classified to obtain the discharge type determination result, including: Wavelet transform is used to decompose the partial discharge signal into multiple scales and extract time-frequency features. The partial discharge signal is decomposed into high-frequency band, mid-frequency band and low-frequency band, and the detail coefficients and approximation coefficients of each frequency band are obtained. Based on the detail coefficients and approximation coefficients, the energy distribution characteristics of the partial discharge signal are obtained. A time-frequency feature vector is constructed based on the energy distribution characteristics. The time-frequency feature vector includes the energy ratio of the high-frequency band to the low-frequency band, the pulse feature function, and the waveform statistical characteristics. The energy ratio of the high-frequency band to the low-frequency band is determined by wavelet coefficients and energy attenuation characteristics. The pulse feature function is composed of amplitude distribution and time characteristics. The waveform statistical characteristics include waveform skewness and waveform kurtosis. Discharge type identification is performed based on the aforementioned time-frequency feature vector, with the following criteria: If the energy ratio of the high frequency band to the low frequency band Greater than the unit value, and the pulse characteristic function If the global maximum value is obtained within the positive half-cycle of the power frequency, and the waveform skewness value is positive, then it is determined to be internal discharge. If the energy ratio of the high frequency band to the low frequency band Less than the unit value, and the pulse characteristic function If the global maximum value is obtained within the negative half-cycle of the power frequency, and the waveform skewness value is negative, then it is determined to be surface discharge. If the pulse characteristic function If multiple peaks occur within one power frequency cycle, and the amplitude ratio of adjacent peaks deviates from the unit value to a degree within a first preset range, while the phase difference of the peaks is greater than the basic segmentation unit of the power frequency cycle, then it is determined to be a floating discharge. If the pulse characteristic function If the multiple continuous peaks appear in the local phase interval, and the peak sequence satisfies the exponential decay law, and the phase difference between adjacent peaks is less than the minimum segmentation unit of the power frequency cycle, it is determined as corona discharge. Based on the above criteria, the discharge type determination result is obtained; Based on the discharge type determination result, the corresponding position correction parameters are determined; The position correction parameters are fused with the initial solution, and the positioning result is continuously corrected through a recursive iterative optimization algorithm until the positioning error converges to within a set threshold, including: The position correction parameters and the initial solution are superimposed and fused in a weighted manner to obtain the initial value for iteration; Construct an error evaluation function for iterative optimization; The iteration parameters are updated based on the calculation results of the error evaluation function. The iteration stops when the relative rate of change of the error evaluation function is less than a preset threshold. Otherwise, the weight coefficient is increased or decreased by comparing the error between two adjacent iterations until the final positioning result is obtained.

2. The method of partial discharge signal detection and location of power distribution cable intermediate joint according to claim 1, characterized in that: The reference point array includes multiple electrode sensors arranged in a ring. Each electrode sensor amplifies and filters the acquired partial discharge signal through a signal conditioning circuit. The signal propagation characteristics include propagation time difference and energy attenuation characteristics.

3. The method of partial discharge signal detection and location of power distribution cable intermediate joint as claimed in claim 2, wherein: Based on the aforementioned signal propagation characteristics, a set of spatial positioning equations for the discharge source is constructed, and the set of spatial positioning equations for the discharge source is solved to obtain an initial solution, including the following steps: Establish a set of positioning equations based on the propagation time difference. Specifically, assume the coordinates of the discharge source are (x0, y0, z0) and the coordinates of the i-th sensor are (xᵢ, yᵢ, zᵢ). Then the signal propagation time difference satisfies: ； in, Let v be the time difference between the arrival of the signal at the i-th sensor and the j-th sensor, and let v be the propagation speed of the electromagnetic wave in the medium. , ∈[1,N] and i≠j, where N is the total number of sensors; The spatial positioning equations of the discharge source are expressed as follows: ； wherein, , are the energy values of the signal reaching the i, jth sensor, respectively, and a is the attenuation coefficient of the medium, , are the distances from the discharge source to the i, jth sensor, respectively. Construct a composite error function: ； wherein m represents a measured value, c represents a calculated value, and is a weight coefficient, is a balance factor; The initial solution is obtained by iterative least squares method; the initial solution represents the initial spatial coordinates of the discharge source.

4. The method for detecting and locating partial discharge signals at intermediate joints of power distribution cables as described in claim 3, characterized in that: The initial solution obtained by the iterative least squares method is expressed by the following formula: ； wherein, denotes the number of iterations, is a Jacobian matrix, is a weight matrix, is a residual vector, is a step factor; Let the weighted residual sum of squares for the first iteration be WSS1=∑i=1nwi(Ri)2 : ； If the weighted residual sum of squares of the current iteration is less than the average of the previous two iterations, i.e.: ； Let the step size factor is represented as: ； Otherwise, let the step size factor be... Represented as: 。 5. The power distribution cable joint partial discharge signal detection and location method of claim 4, wherein: Based on the discharge type determination result, the corresponding position correction parameters are determined, including the following steps: A discharge location feature model is constructed; the discharge location feature model includes a location feature parameter matrix for different discharge types, and the location feature parameter matrix is ​​established based on the high-low frequency band energy ratio, pulse feature function, and waveform statistical features. Establish coordinate transformation relationships between the rectangular coordinate system and the cylindrical coordinate system; Constructing a base location feature function : ； in, For location feature functions, This is the normalized value of the energy ratio between high and low frequency bands. The normalized impulse characteristic function, For waveform skewness, For waveform kurtosis, , , The feature weight index; , , These are the spatial position coordinates in a cylindrical coordinate system. Radial distance, It is the azimuth angle. This refers to the axial height. If the discharge type determination result is the internal discharge and surface discharge type, a radial dominant type characteristic function is constructed : ； wherein is a radial decay coefficient; is a radial correction factor; If the discharge type determination result is the floating discharge type, a balanced type characteristic function is constructed : ； wherein is a radial characteristic dimension, is an angular characteristic dimension, is an axial characteristic dimension, is an equalization factor; If the discharge type determination result is the corona discharge type, an axial-dominant characteristic function is constructed : ； wherein, is an angular modulation coefficient; is a height modulation coefficient; is an axial correction factor; Extract the correction parameter matrix and transform it from cylindrical coordinates to Cartesian coordinates; The position correction parameters are calculated using matrix multiplication based on the correction parameter matrix in the rectangular coordinate system and the spatial coordinate values ​​to be corrected.

6. The partial discharge signal detection and location system for power cable joint, based on the partial discharge signal detection and location method for power cable joint according to any one of claims 1-5, characterized in that: include, The signal acquisition module is used to acquire the signal propagation characteristics of partial discharge signals by means of an array of reference points set around the intermediate joint of the power distribution cable; The spatial positioning module is used to construct a set of spatial positioning equations for the discharge source based on the signal propagation characteristics, and to solve the set of spatial positioning equations for the discharge source to obtain an initial solution. The pattern recognition module is used to analyze the time-frequency characteristics of the partial discharge signal, identify and classify the partial discharge signal, and obtain the discharge type determination result. The parameter extraction module determines the corresponding position correction parameters based on the discharge type determination result; The optimization iteration module is used to fuse the position correction parameters with the initial solution and continuously correct the positioning result through a recursive iterative optimization algorithm until the positioning error converges to within a set threshold. 7.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is configured to perform the method according to any one of claims 1-6 when the computer program is executed by the processor. When the processor executes the computer program, it implements the steps of the method for detecting and locating partial discharge signals of intermediate joints of power distribution cables as described in any one of claims 1 to 5.

8. A computer-readable storage medium having stored thereon a computer program, characterized in that When the computer program is executed by the processor, it implements the steps of the method for detecting and locating partial discharge signals of intermediate joints of power distribution cables as described in any one of claims 1 to 5.

Images