Local discharge source positioning method and device and computer program product

Abstract

The invention discloses a partial discharge source positioning method and device and a computer program product, and the method comprises the steps: S1, converting a sensor site error into an equivalent time error based on Monte Carlo simulation, and constructing a comprehensive time error model in combination with a time-of-arrival measurement error; s2, correcting the time of arrival based on the comprehensive time error model, establishing a nonlinear TDOA equation set of a sensor array by using a time-of-arrival method, and resolving three-dimensional coordinates of a partial discharge source through a hierarchical optimization algorithm; and S3, for a multi-discharge source scene, generating a discharge source coordinate set through Monte Carlo simulation, performing density clustering analysis on the discharge source coordinate set to separate independent signal sources, and outputting positioning coordinates of each independent signal source. According to the method, the precision and reliability of partial discharge positioning of the high-voltage switch cabinet are remarkably improved, and effective technical support is provided for insulation state evaluation and fault early warning of power equipment.

Description

Technical Field

[0001] The present invention relates to the technical field of power systems, and in particular to a method and device for locating a local discharge source, and a computer program product. Background Art

[0002] Partial discharge is an important phenomenon that indicates insulation degradation in high-voltage switchgear. Its precise positioning plays a key role in equipment status assessment and fault prevention. Existing positioning systems based on the Time Difference of Arrival (TDOA) method have significant drawbacks. First, there is a complex coupling effect between the site error caused by sensor installation position deviation, the timing error generated by the arrival time extraction process, and the measurement error of the medium sound velocity parameter. These multiple errors are superimposed and propagated in the positioning calculation, causing the final positioning result to deviate from the actual discharge source location. Second, the traditional Monte Carlo simulation method requires the simultaneous processing of random variables from multiple error sources. The calculation process involves massive iterative operations, and a single simulation takes tens of thousands of seconds, which is difficult to meet the real-time requirements of online monitoring of power equipment. Third, under complex operating conditions with multiple discharge sources, the existing algorithm lacks an effective signal separation and feature identification mechanism, resulting in mutual interference between the signals of different discharge sources, confusion in the positioning results, and serious lack of adaptability of the system to multi-source scenarios. Finally, the traditional Monte Carlo simulation method does not establish a clear error propagation mathematical model, and the random parameter generation process lacks a theoretical basis. This leads to significant differences between the simulation results and the actual error distribution, affecting the credibility of the positioning reliability assessment. These technical bottlenecks have restricted the engineering application of partial discharge positioning technology for high-voltage equipment. Summary of the Invention

[0003] The technical problem to be solved by the embodiments of the present invention is to provide a method, an apparatus and a computer program product for locating a partial discharge source, so as to improve the positioning accuracy of the partial discharge source.

[0004] To solve the above technical problems, the present invention provides a method for locating a local discharge source, comprising:

[0005] Step S1, converting the sensor site error into an equivalent time error based on Monte Carlo simulation, and building a comprehensive time error model in combination with the arrival time measurement error;

[0006] Step S2, correcting the arrival time based on the integrated time error model, establishing a nonlinear TDOA equation group for the sensor array using the time difference of arrival method, and calculating the three-dimensional coordinates of the partial discharge source using a hierarchical optimization algorithm;

[0007] Step S3: for a multi-discharge source scenario, a discharge source coordinate set is generated through Monte Carlo simulation, a density cluster analysis is performed on the discharge source coordinate set to separate independent signal sources, and the positioning coordinates of each independent signal source are output.

[0008] Preferably, in step S1, the conversion formula for converting the sensor site error to the time error is:

[0009]

[0010] Where ΔL is the distance deviation between the actual sensor installation position and the theoretical installation position with respect to the discharge source, (x, y, z) are the actual location coordinates of the local discharge source, (x1, y1, z1) are the theoretical installation position coordinates of sensor 1, Δx1, Δy1, and Δz1 are the three mutually perpendicular components of the site error, c is the sound velocity in the discharge source space, and L1 is the theoretical distance between sensor 1 and the discharge source.

[0011] Preferably, in step S2, the hierarchical optimization algorithm specifically includes:

[0012] Decompose the three-dimensional coordinate solution into the analytical solution of plane coordinates x and y and the traversal optimization of the vertical coordinate z;

[0013] The nonlinear equations are simplified into linear equations in plane coordinates x and y by elimination method and solved;

[0014] Fixed the plane coordinates x and y, traversed the search space of the vertical coordinate z, and determined the optimal z coordinate with the goal of minimizing the sum of squares of the arrival time residuals.

[0015] Preferably, in step S2, when traversing the search space of the vertical coordinate z, a grid traversal algorithm with a preset accuracy of 1 cm is adopted.

[0016] Preferably, in step S3, the density cluster analysis adopts DBSCAN algorithm to analyze each positioning point (x i ,y i ,z i ), calculate its Euclidean distance to other points as follows:

[0017]

[0018] If d ij ≤ε,(x j ,y j ,z j ) is marked as (x i ,y i ,z i)’s neighborhood points; if the number of samples in a point’s neighborhood is ≥ MinPts, it is marked as a core point; starting from the core point, all reachable points in its neighborhood are recursively merged to form an independent cluster; where ε is the spatial distance threshold and MinPts is the minimum number of samples.

[0019] Preferably, in step S3, the output positioning coordinates of each independent signal source are the geometric mean of all coordinate points in the corresponding cluster, and the calculation formula is:

[0020]

[0021] Where N is the number of samples in the cluster.

[0022] Preferably, the spatial distance threshold ε is 1 meter, and the minimum number of samples MinPts is 5.

[0023] The present invention also provides a local discharge source positioning device, comprising:

[0024] The error modeling module is used to convert the sensor site error into an equivalent time error based on Monte Carlo simulation, and to construct a comprehensive time error model by combining the arrival time measurement error;

[0025] A coordinate calculation module, configured to correct the arrival time based on the integrated time error model, establish a nonlinear TDOA equation group for the sensor array using the time difference of arrival method, and solve the three-dimensional coordinates of the local discharge source using a hierarchical optimization algorithm;

[0026] The multi-source clustering module generates a set of discharge source coordinates through Monte Carlo simulation for multiple discharge source scenarios, performs density clustering analysis on the set of discharge source coordinates to separate independent signal sources, and outputs the positioning coordinates of each independent signal source.

[0027] The present invention also provides a local discharge source positioning device, comprising:

[0028] one or more processors;

[0029] Memory;

[0030] One or more application programs, wherein the one or more application programs are stored in the memory and configured to be executed by the one or more processors, and the one or more application programs are configured to perform the partial discharge source locating method.

[0031] The present invention also provides a computer program product, comprising computer instructions, wherein the computer instructions instruct a computer device to execute operations corresponding to the method.

[0032] The implementation of the present invention has the following beneficial effects: the present invention uses Monte Carlo simulation to equate the sensor site error to a time error, and combines the wave arrival time measurement error to construct a comprehensive time error model, effectively decoupling the multiple coupling effects of site deviation, timing error, and sound speed error in traditional methods, greatly improving the convergence speed of positioning error; using a hierarchical optimization algorithm to decompose the three-dimensional coordinate solution into a two-stage process of plane coordinate analytical solution and vertical coordinate traversal optimization, using the elimination method to establish a plane coordinate linear equation system for fast solution, and then optimizing the z coordinate with a 1cm precision grid search. Compared with the traditional global search method, the computational efficiency is significantly improved, meeting the real-time requirements of online monitoring; for multi-source discharge scenarios, the present invention generates a coordinate set through Monte Carlo simulation and integrates density cluster analysis, combines wavelet threshold denoising and DBSCAN algorithm to achieve independent cluster separation and noise point filtering of multi-source signals, and solves the positioning confusion problem of traditional algorithms under multi-source working conditions. The present invention significantly improves the accuracy and reliability of localization of partial discharge in high-voltage switchgear, and provides effective technical support for insulation status assessment and fault warning of power equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0034] Figure 1 It is a flow chart of a method for locating a local discharge source according to an embodiment of the present invention.

[0035] Figure 2 This is a schematic diagram of a specific flow chart of a method for locating a local discharge source according to an embodiment of the present invention. DETAILED DESCRIPTION

[0036] The following descriptions of the embodiments refer to the accompanying drawings to illustrate specific embodiments in which the present invention may be implemented.

[0037] Please refer to Figure 1 As shown, a first embodiment of the present invention provides a method for locating a local discharge source, comprising:

[0038] Step S1, converting the sensor site error into an equivalent time error based on Monte Carlo simulation, and building a comprehensive time error model in combination with the arrival time measurement error;

[0039] Step S2, correcting the arrival time based on the integrated time error model, establishing a nonlinear TDOA equation group for the sensor array using the time difference of arrival method, and solving the three-dimensional coordinates of the local discharge source using a hierarchical optimization algorithm;

[0040] Step S3: for a multi-discharge source scenario, a discharge source coordinate set is generated through Monte Carlo simulation, a density cluster analysis is performed on the discharge source coordinate set to separate independent signal sources, and the positioning coordinates of each independent signal source are output.

[0041] Specifically, please combine Figure 2 As shown, the present invention provides a local discharge source location method based on the arrival time difference method and Monte Carlo simulation design. The station site error is equivalent to the time error through Monte Carlo simulation design, and a hierarchical optimization algorithm is used to preferentially solve the x and y coordinates and then traverse the z coordinate space to achieve accurate positioning of the discharge source.

[0042] Step S1 simplifies the error model by converting the site error (Δx) into an equivalent time error (Δt) in the following way. This allows the site error to be ignored during the Monte Carlo simulation, and only the time error to be considered, thus reducing the computational effort:

[0043]

[0044] Where ΔL is the distance deviation between the actual sensor installation position and the theoretical installation position with respect to the discharge source, (x, y, z) are the actual location coordinates of the local discharge source, (x1, y1, z1) are the theoretical installation position coordinates of sensor 1, Δx1, Δy1, and Δz1 are the three mutually perpendicular components of the site error, c is the sound velocity in the discharge source space, and L1 is the theoretical distance from sensor 1 to the discharge source, that is, the straight-line geometric distance from sensor 1 to the local discharge source under the ideal assumption that the sensor installation position is completely accurate (no site error) and the arrival time measurement is error-free.

[0045] Because the sensor array is fixed, Monte Carlo simulations can be used to analyze the positioning capabilities of different algorithms. As long as the sound source coordinates used in the simulations are the same, the simulation results are comparable, and the site location errors can be ignored. Monte Carlo simulations only need to consider time errors, reducing the computational complexity by over 80%.

[0046] Step S2 executes the hierarchical optimization algorithm, which specifically includes:

[0047] (a) Decompose the three-dimensional coordinate solution into the analytical solution of the plane coordinates x and y and the traversal optimization of the vertical coordinate z;

[0048] (b) Simplify the nonlinear equations into a linear equation system in plane coordinates x and y by elimination and solve them;

[0049] (c) Fixing the plane coordinates x and y, traverse the search space of the vertical coordinate z and determine the optimal z coordinate by minimizing the sum of squared arrival time residuals.

[0050] Furthermore, the plane coordinates are quickly solved:

[0051] Assume there are three sensors located at (x1, y1, z1), (x2, y2, z2), and (x3, y3, z3). Based on the time difference of arrival (TDOA), the distance difference equation can be established:

[0052]

[0053] Where c is the speed of sound in the discharge source space, Δt ij is the time difference between sensors i and j.

[0054] The TDOA quadratic equations are simplified to a linear equation of x,y coordinates by elimination method. Take the TDOA equations of sensors 1, 2 and sensors 1, 3 as examples:

[0055] Change the equation:

[0056]

[0057] After subtraction, square and expansion, z is eliminated 2 Term, and finally get the linear equation:

[0058]

[0059] It can be expressed as:

[0060] a1x+b1y=c1

[0061] Then, through the TDOA equations of sensors 1 and 2 and sensors 2 and 3, we can obtain:

[0062] a²x+b²y=c²

[0063] Combining x and y, we can obtain a system of linear equations in two variables, and then prioritize solving the plane coordinates.

[0064] After fixing x and y, perform a 1cm precision grid traversal on the z coordinate to avoid the complexity of a 3D global search. The optimal z value is determined by minimizing the sum of squares of the arrival time residuals:

[0065]

[0066] Among them, T i The boda time.

[0067] The embodiment of the present invention uses a hierarchical optimization algorithm to independently locate the multi-source discharge signal and generate a coordinate data set S = {(x1, y1, z1), (x2, y2, z2), ..., (x n ,y n ,z nWavelet threshold denoising is applied to the dataset to eliminate abnormal positioning points and remove invalid data that are obviously beyond the physical space range.

[0068] Combined with the DBSCAN clustering algorithm analysis, the DBSCAN clustering algorithm parameters are set as follows: ε (spatial distance threshold): 1 meter; MinPts (minimum number of samples): 5 (parameters are adjusted according to the actual confusion rate). i ,y i ,z i ), calculate its Euclidean distance to other points:

[0069]

[0070] If d ij ≤ε,(x j ,y j ,z j ) is marked as (x i ,y i ,z i ) neighborhood points. If the number of samples in a point's neighborhood is ≥ MinPts (i.e., at least 5 points are within a 1-meter range), it is marked as a core point. Starting from the core point, all reachable points in its neighborhood are recursively merged to form an independent cluster. Isolated points that cannot be included in any cluster are considered noise and excluded from the multi-source analysis. Each cluster corresponds to an independent discharge source; the geometric mean of all points in the cluster is taken as the final location coordinate of the discharge source:

[0071]

[0072] Where N is the number of samples in the cluster.

[0073] In this way, it is finally possible to distinguish multi-source signals and avoid cross interference.

[0074] When implementing the present invention, the hardware deployment is as follows:

[0075] A circular four-element ultrasonic array with a radius of 35mm is arranged in the switch cabinet, with the center of the array coinciding with the origin of the switch cabinet coordinate system. The sensor elements are installed on the side walls of the cabinet to ensure coverage of areas with high incidence of partial discharge. After the sensor elements are installed, they are calibrated using a laser rangefinder to ensure that the site error Δx is ≤ 1mm.

[0076] Then, the method for locating the local discharge source according to the embodiment of the present invention is implemented, including the following steps:

[0077] Data collection: collect the arrival time of each sensor and use wavelet threshold denoising to remove interference (such as wavelet threshold denoising);

[0078] TDOA modeling: Establish a TDOA quadratic equation system, eliminate the z term by pairwise subtraction, and obtain the linear equations of x and y;

[0079] Layered optimization: Select the formula with the smallest error to solve the plane position; after fixing x and y, traverse the z coordinate (1000-5000mm) with a step size of 1cm, calculate the objective function value, and select the z coordinate corresponding to the minimum value;

[0080] Multi-source expansion: Perform the above steps independently for multi-source signals and distinguish different discharge sources through cluster analysis (spatial distance threshold ≤ 1m).

[0081] Verification and optimization:

[0082] Since the array parameters are fixed, 10,000 simulations are performed on the expressions of x and y respectively. After statistically analyzing the distribution of the absolute errors of x and y, the performance of different algorithms is compared to verify the comprehensive advantages of the hierarchical optimization algorithm in terms of accuracy and efficiency.

[0083] Table 1 Description of the four algorithms

[0084]

[0085] Table 2 Comparison of four methods

[0086]

[0087] As shown in Tables 1 and 2, Method 2, by traversing the entire space, is able to obtain the global optimal solution as much as possible. Therefore, its error statistics are superior to those of the other methods. However, its computation time is much longer than that of the other algorithms, and it cannot meet the timeliness requirements of the online monitoring system. Method 3's error statistics are second only to Method 2, and its computation speed is second only to Method 4, which can simultaneously meet the error and timeliness requirements of partial discharge location. The particle swarm algorithm (Method 4) is used to iteratively solve the optimization function. Although the overall error (median) is small, the optimization function has many local optimal points in the presence of time errors. The iterative solution is prone to falling into local optimal points, and large errors are easily incurred during the solution process.

[0088] Compared with other methods, the method 3 (solving the linear equation to obtain x and y, and obtaining z by traversing the optimization function space) obtained by the embodiment of the present invention through Monte Carlo simulation optimization has better accuracy and timeliness.

[0089] The local discharge source location method according to the embodiment of the present invention is widely used, for example:

[0090] In the field of high-voltage equipment monitoring, it is suitable for online monitoring and positioning of partial discharge in switchgear, transformers, GIS and other equipment;

[0091] In the field of intelligent operation and maintenance systems, integration into the power Internet of Things platform can achieve real-time early warning and precise positioning of insulation defects;

[0092] In the field of multi-source fault diagnosis, it supports the positioning and differentiation of concurrent discharge scenarios of multiple devices in substations, improving operation and maintenance efficiency.

[0093] Corresponding to the method for locating a local discharge source described in the first embodiment of the present invention, a second embodiment of the present invention provides a device for locating a local discharge source, comprising:

[0094] The error modeling module is used to convert the sensor site error into an equivalent time error based on Monte Carlo simulation, and to construct a comprehensive time error model by combining the arrival time measurement error;

[0095] A coordinate calculation module, configured to correct the arrival time based on the integrated time error model, establish a nonlinear TDOA equation group for the sensor array using the time difference of arrival method, and solve the three-dimensional coordinates of the local discharge source using a hierarchical optimization algorithm;

[0096] The multi-source clustering module generates a set of discharge source coordinates through Monte Carlo simulation for multiple discharge source scenarios, performs density clustering analysis on the set of discharge source coordinates to separate independent signal sources, and outputs the positioning coordinates of each independent signal source.

[0097] Corresponding to the method for locating a local discharge source described in the first embodiment of the present invention, the third embodiment of the present invention further provides a device for locating a local discharge source, comprising:

[0098] one or more processors;

[0099] Memory;

[0100] One or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the one or more processors, and the one or more applications are configured to execute the local discharge source locating method described in the first embodiment of the present invention.

[0101] Corresponding to the local discharge source locating method described in the first embodiment of the present invention, the fourth embodiment of the present invention further provides a computer program product, including computer instructions, which instruct a computer device to perform operations corresponding to the local discharge source locating method described in the first embodiment of the present invention.

[0102] Preferably, the processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor, or the processor may be any conventional processor. The processor is the control center of the device, and various parts of the device are connected using various interfaces and lines.

[0103] The memory mainly includes a program storage area and a data storage area, wherein the program storage area can store an operating system, an application program required for at least one function, etc., and the data storage area can store related data, etc. In addition, the memory can be a high-speed random access memory, or a non-volatile memory, such as a plug-in hard disk, a smart memory card (SmartMedia Card, SMC), a secure digital (Secure Digital, SD) card, and a flash card, etc., or the memory can also be other volatile solid-state storage devices.

[0104] It should be noted that the above-mentioned device may include but is not limited to a processor and a memory, which can be understood by those skilled in the art.

[0105] From the above description, it can be seen that compared with the existing technology, the beneficial effects of the present invention are as follows: the present invention uses Monte Carlo simulation to equate the sensor site error to time error, and combines the wave arrival time measurement error to construct a comprehensive time error model, effectively decoupling the multiple coupling effects of site deviation, timing error and sound speed error in the traditional method, which greatly improves the convergence speed of positioning error; adopts a hierarchical optimization algorithm to decompose the three-dimensional coordinate solution into a two-stage process of plane coordinate analytical solution and vertical coordinate traversal optimization, uses the elimination method to establish a plane coordinate linear equation system for fast solution, and optimizes the z coordinate with a 1cm precision grid search. Compared with the traditional global search method, the computational efficiency is significantly improved, meeting the real-time requirements of online monitoring; for the multi-source discharge scenario, the coordinate set is generated by Monte Carlo simulation and integrated with density cluster analysis, combined with wavelet threshold denoising and DBSCAN algorithm, to achieve independent cluster separation and noise point filtering of multi-source signals, solving the positioning confusion problem of traditional algorithms under multi-source working conditions. The present invention significantly improves the accuracy and reliability of partial discharge positioning in high-voltage switchgear, providing effective technical support for insulation status assessment and fault warning of power equipment.

[0106] The above disclosure is merely a preferred embodiment of the present invention and is not intended to limit the scope of the present invention. Therefore, equivalent changes made according to the claims of the present invention are still within the scope of the present invention.

Claims

1. A method for locating a local discharge source, characterized in that: include: Step S1, converting the sensor site error into an equivalent time error based on Monte Carlo simulation, and building a comprehensive time error model in combination with the arrival time measurement error; Step S2, correcting the arrival time based on the integrated time error model, establishing a nonlinear TDOA equation group for the sensor array using the time difference of arrival method, and solving the three-dimensional coordinates of the local discharge source using a hierarchical optimization algorithm; Step S3: for a multi-discharge source scenario, a discharge source coordinate set is generated through Monte Carlo simulation, a density cluster analysis is performed on the discharge source coordinate set to separate independent signal sources, and the positioning coordinates of each independent signal source are output.

2. The method according to claim 1, characterized in that In step S1, the conversion formula for converting the sensor site error to the time error is: Where ΔL is the distance deviation between the actual sensor installation position and the theoretical installation position with respect to the discharge source, (x, y, z) are the actual location coordinates of the local discharge source, (x1, y1, z1) are the theoretical installation position coordinates of sensor 1, Δx1, Δy1, and Δz1 are the three mutually perpendicular components of the site error, c is the sound velocity in the discharge source space, and L1 is the theoretical distance between sensor 1 and the discharge source.

3. The method according to claim 1, characterized in that In step S2, the hierarchical optimization algorithm specifically includes: Decompose the three-dimensional coordinate solution into the analytical solution of plane coordinates x and y and the traversal optimization of the vertical coordinate z; The nonlinear equations are simplified into linear equations in plane coordinates x and y by elimination method and solved; Fixed the plane coordinates x and y, traversed the search space of the vertical coordinate z, and determined the optimal z coordinate with the goal of minimizing the sum of squares of the arrival time residuals.

4. The method according to claim 3, characterized in that In the step S2, when traversing the search space of the vertical coordinate z, a grid traversal algorithm with a preset accuracy of 1 cm is adopted.

5. The method according to claim 1, wherein In step S3, the density cluster analysis uses the DBSCAN algorithm to perform clustering on each location point (x i ,y i ,z i ), calculate its Euclidean distance to other points as follows: If d ij ≤ε,(x j ,y j ,z j ) is marked as (x i ,y i ,z i )’s neighborhood points; if the number of samples in a point’s neighborhood is ≥ MinPts, it is marked as a core point; starting from the core point, all reachable points in its neighborhood are recursively merged to form an independent cluster; where ε is the spatial distance threshold and MinPts is the minimum number of samples.

6. The method according to claim 5, characterized in that In step S3, the output positioning coordinates of each independent signal source are the geometric mean of all coordinate points in the corresponding cluster, and the calculation formula is: Where N is the number of samples in the cluster.

7. The method according to claim 5, characterized in that The spatial distance threshold ε is 1 meter, and the minimum number of samples MinPts is 5.

8. A local discharge source positioning device, characterized in that: include: The error modeling module is used to convert the sensor site error into an equivalent time error based on Monte Carlo simulation, and to construct a comprehensive time error model by combining the arrival time measurement error; A coordinate calculation module, configured to correct the arrival time based on the integrated time error model, establish a nonlinear TDOA equation group for the sensor array using the time difference of arrival method, and calculate the three-dimensional coordinates of the partial discharge source using a hierarchical optimization algorithm; The multi-source clustering module generates a set of discharge source coordinates through Monte Carlo simulation for multiple discharge source scenarios, performs density clustering analysis on the set of discharge source coordinates to separate independent signal sources, and outputs the positioning coordinates of each independent signal source.

9. A local discharge source positioning device, characterized in that: include: one or more processors; Memory; One or more application programs, wherein the one or more application programs are stored in the memory and configured to be executed by the one or more processors, and the one or more application programs are configured to execute the partial discharge source locating method according to any one of claims 1 to 7.

10. A computer program product, characterized in that The method comprises computer instructions, wherein the computer instructions instruct a computer device to perform operations corresponding to the method according to any one of claims 1 to 7.

Images