Circuit breaker mechanical fault analysis method and system based on dynamic time warping method

Abstract

The present invention relates to the technical field of power fault analysis, and discloses a circuit breaker mechanical fault analysis method and system based on a dynamic time warping method. The dynamic time warping method is used to warp multiple characteristic operation reference time series data of the circuit breaker under normal operation to obtain an optimal warping path reference curve about the time series, thereby intuitively characterizing the operation trend of the circuit breaker under normal operation. The characteristic operation time series path curve of the current operation period of the circuit breaker to be tested is judged by the similarity between the optimal warping path reference curve and the characteristic operation time series path curve of the current operation period of the circuit breaker to be tested, thereby judging the fault state of the circuit breaker to be tested, thereby making it possible to estimate the fault state of the circuit breaker to be tested in real time, and improving the accuracy of the fault state assessment of the circuit breaker.

Description

Technical Field

[0001] The present invention relates to the technical field of power fault analysis, and in particular to a circuit breaker mechanical fault analysis method and system based on a dynamic time warping method. Background Art

[0002] Circuit breakers are crucial control devices in the power distribution system, ensuring continuity and stability of the system. Circuit breakers operate under strong magnetic fields, high currents, and high voltages for extended periods, making monitoring and protecting them throughout their normal operating lifecycle crucial. Therefore, effectively monitoring changes in mechanical characteristics, identifying potential faults in advance, and promptly eliminating them are key to reducing circuit breaker failure rates, minimizing power outages, and lowering maintenance costs.

[0003] Once a circuit breaker's mechanical characteristics deviate from the permitted range, they will directly affect the safety and stability of the circuit breaker's operation. However, it is currently difficult to timely and accurately assess the mechanical characteristics of circuit breakers, making it difficult to reasonably schedule maintenance. Summary of the Invention

[0004] The present invention provides a circuit breaker mechanical fault analysis method and system based on the dynamic time warping method, which solves the current technical problem that it is difficult to timely and accurately evaluate the mechanical characteristic operation faults of the circuit breaker, resulting in difficulty in reasonably arranging maintenance time.

[0005] In view of this, a first aspect of the present invention provides a circuit breaker mechanical fault analysis method based on a dynamic time warping method, comprising:

[0006] Acquire multiple characteristic operation benchmark time series data of the circuit breaker under normal operation, where the data types of the multiple characteristic operation benchmark time series data are the same;

[0007] Preprocessing a plurality of characteristic operation benchmark time series data;

[0008] Performing warping processing on the plurality of characteristic operation benchmark time series data based on a dynamic time warping method to determine an optimal warping path benchmark curve for the plurality of characteristic operation benchmark time series data;

[0009] Performing numerical fitting on characteristic operation timing data of the current operation period of the circuit breaker to be tested to obtain a characteristic operation timing path curve of the current operation period of the circuit breaker to be tested;

[0010] Comparing the similarity between the characteristic operation sequence path curve of the current operation period of the circuit breaker to be tested and the optimal regularized path reference curve;

[0011] When the similarity is less than a preset similarity threshold, it is determined that the current operating period of the circuit breaker to be tested is in a fault state.

[0012] Optionally, the data type of the characteristic operation reference timing data is opening and closing stroke timing data or opening and closing coil current timing data.

[0013] Optionally, the preprocessing includes weighted filtering.

[0014] Optionally, the step of performing warping processing on the plurality of characteristic operation benchmark time series data based on the dynamic time warping method to determine the optimal warping path reference curve of the plurality of characteristic operation benchmark time series data specifically includes:

[0015] Constructing multiple n×m sparse matrices of the characteristic operation benchmark time series data, where n represents the number of columns of cells in the sparse matrix, and m represents the number of rows of cells in the sparse matrix;

[0016] Initializing cells in the sparse matrix to empty values;

[0017] Normalizing the plurality of characteristic operation benchmark time series data;

[0018] Calculate the mean of all normalized feature-based benchmark time series data to obtain a total mean;

[0019] Performing difference operations on multiple normalized feature operation benchmark time series data and the total mean, filling values ​​in corresponding cells in the sparse matrix according to the differences, and forming a search path range;

[0020] Within the search path range, starting from the cells with non-empty values ​​in the bottom row of the sparse matrix, traverse upward in sequence until reaching the cells with non-empty values ​​in the top row of the sparse matrix, to obtain multiple search paths and cumulative distance values ​​corresponding to the multiple search paths;

[0021] A search path with the smallest cumulative distance value is screened out, and curve fitting is performed on the screened search path to obtain an optimal regularized path reference curve of the plurality of characteristic operation reference time series data.

[0022] Optionally, the step of performing difference operations on the plurality of normalized characteristic operation benchmark time series data and the total mean, filling values ​​in corresponding cells in the sparse matrix according to the differences, and forming a search path range specifically includes:

[0023] Performing difference calculations on the plurality of normalized characteristic operation benchmark time series data and the total mean value, and determining whether the difference is less than a preset difference threshold;

[0024] If it is determined that the difference is less than the preset difference threshold, updating the element of the cell corresponding to the normalized feature operation benchmark time series data in the sparse matrix to the difference;

[0025] If it is determined that the difference is not less than the preset difference threshold, the element of the cell corresponding to the normalized feature operation benchmark time series data in the sparse matrix remains a null value;

[0026] Filter out cells corresponding to non-empty values ​​of the sparse matrix to form an initial search path range;

[0027] Determine whether there is an isolated cell in the sparse matrix, and the principle for determining the isolated cell is: the adjacent cells of the cell are all null values;

[0028] If it is determined that the isolated cell exists in the sparse matrix, the adjacent cells corresponding to the isolated cell are filled with values, and the values ​​filled in the adjacent cells are the differences between the normalized feature operation benchmark time series data corresponding to the cell and the total mean;

[0029] The initial search path range is updated according to the cells filled with values ​​to form a search path range.

[0030] Optionally, the method further comprises:

[0031] Calculating the Euclidean distance between each two adjacent cells according to the positions of each two adjacent cells in the search path;

[0032] All Euclidean distances in the search path are accumulated to obtain a cumulative distance value of the search path.

[0033] In a second aspect, the present invention further provides a circuit breaker mechanical fault analysis system based on a dynamic time warping method, comprising:

[0034] A reference data acquisition module is used to acquire a plurality of characteristic operation reference time series data of the circuit breaker under normal operation, wherein the data types of the plurality of characteristic operation reference time series data are the same;

[0035] A preprocessing module, configured to preprocess the plurality of characteristic operation benchmark time series data;

[0036] A time series warping module is used to warp the plurality of characteristic operation benchmark time series data based on a dynamic time warping method, and determine an optimal warping path reference curve for the plurality of characteristic operation benchmark time series data;

[0037] A numerical fitting module is used to perform numerical fitting on characteristic operation sequence data of the current operation period of the circuit breaker to be tested, so as to obtain a characteristic operation sequence path curve of the current operation period of the circuit breaker to be tested;

[0038] A similarity module, configured to compare the similarity between the characteristic operation sequence path curve of the current operation period of the circuit breaker to be tested and the optimal regularized path reference curve;

[0039] The fault judgment module is configured to judge that the current operation period of the circuit breaker to be tested is in a fault state when the similarity is less than a preset similarity threshold.

[0040] In a third aspect, the present invention also provides an electronic device comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the circuit breaker mechanical fault analysis method based on the dynamic time warping method as described above.

[0041] In a fourth aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed, implements the steps of the circuit breaker mechanical fault analysis method based on the dynamic time warping method as described above.

[0042] In a fifth aspect, the present invention also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions, wherein when the program instructions are executed by a computer, the computer executes the steps of the circuit breaker mechanical fault analysis method based on the dynamic time warping method as described above.

[0043] It can be seen from the above technical solutions that the present invention has the following advantages:

[0044] The present invention uses a dynamic time warping method to warp multiple characteristic operating benchmark time series data of the circuit breaker under normal operation, and obtains the optimal warping path benchmark curve about the time series, so as to intuitively characterize the operating trend of the circuit breaker under normal operation, and judge the degree of similarity between the characteristic operating sequence path curve of the current operating period of the circuit breaker to be tested and the optimal warping path benchmark curve, so as to judge the fault state of the circuit breaker to be tested, so that the fault state of the circuit breaker to be tested can be estimated in real time, and the accuracy of the fault state assessment of the circuit breaker is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 A flow chart of a circuit breaker mechanical fault analysis method based on dynamic time warping method provided by an embodiment of the present invention;

[0046] Figure 2A schematic diagram of a sparse matrix during initialization provided by an embodiment of the present invention;

[0047] Figure 3 A schematic diagram of the initial search path range of a sparse matrix provided by an embodiment of the present invention;

[0048] Figure 4 A schematic diagram of an optimal regularized path reference curve provided by an embodiment of the present invention;

[0049] Figure 5 A schematic structural diagram of a circuit breaker mechanical fault analysis system based on a dynamic time warping method provided by an embodiment of the present invention;

[0050] Figure 6 A schematic structural diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0051] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0052] It should be noted that the terms "including" and "having" and any variations thereof in the specification and claims of the present invention and the above-mentioned drawings are intended to cover non-exclusive inclusions. For example, a process, method, system, product or apparatus comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or are inherent to these processes, methods, products or apparatuses.

[0053] This embodiment is applicable to the analysis of mechanical failures of circuit breakers. The method can be performed by a circuit breaker mechanical failure analysis device. The circuit breaker mechanical failure analysis device can be implemented in the form of hardware and / or software, and the circuit breaker mechanical failure analysis device can be configured in a computer device.

[0054] like Figure 1 As shown, Figure 1 The flow chart of a circuit breaker mechanical fault analysis method based on dynamic time warping method provided by the present invention is illustrated.

[0055] The present invention provides a circuit breaker mechanical fault analysis method based on dynamic time warping method, comprising:

[0056] Step S1: Acquire multiple characteristic operation benchmark time series data of a circuit breaker under normal operation, where the data types of the multiple characteristic operation benchmark time series data are the same.

[0057] The data type of the characteristic operation benchmark timing data is opening and closing stroke timing data or opening and closing coil current timing data.

[0058] Specifically, the displacement timing signals or coil current timing signals generated during the normal opening and closing of the circuit breaker can be collected using relevant sensor devices and timers. Closing stroke timing data or opening and closing coil current timing data can be generated using the displacement timing signals or coil current timing signals. The frequency and length of the collection of multiple characteristic operating benchmark timing data are consistent.

[0059] It can be understood that the plurality of characteristic operation reference time series data of the circuit breaker under normal operation should be the characteristic operation reference time series data of the circuit breaker under no fault condition.

[0060] Step S2: Preprocess multiple feature operation benchmark time series data.

[0061] Understandably, circuit breakers may be subject to various interferences, such as the operation of other mechanical equipment in the surrounding environment, electromagnetic interference from electrical equipment, and natural factors such as wind and temperature fluctuations, all of which can generate vibration noise. This noise can be superimposed on the target vibration signal, affecting its purity. Furthermore, sensor non-idealities, such as sensitivity drift, noise figure, and frequency response limitations, can cause distortion in the acquired signal. For example, the narrow bandwidth of a velocity sensor may not effectively cover the entire vibration frequency range of interest. During the signal transmission process from the sensor to the data acquisition system, electromagnetic interference from cables, poor grounding, or signal attenuation can also introduce interference. Furthermore, due to the mechanical coupling effect of the circuit breaker, mechanical coupling between the mechanism itself and other components, such as resonance and the clattering of loose parts, can generate additional vibration signals, which can be unintended sources of interference. Furthermore, for large equipment installed on the ground, ground vibrations can also be transmitted to the operating mechanism, acting as a source of interference.

[0062] Regarding the current signal collected in the experiment, the coil, as an electromagnetic element, generates a magnetic field when energized and is also affected by the surrounding electromagnetic fields. Surrounding electronic equipment, high-voltage lines, inverters, and other devices can generate strong electromagnetic radiation, which can couple into the measurement circuit, causing waveform distortion or noise, resulting in glitches and jitter. Compared to current and vibration signals, the displacement signal is significantly less susceptible to interference. Displacement signals are typically a physical quantity that directly reflects the position change of mechanical components. They have a low signal frequency and are relatively smooth and stable in the absence of severe external vibration. This low-frequency characteristic makes them less susceptible to high-frequency electromagnetic interference.

[0063] To this end, when preprocessing a plurality of characteristic operation benchmark time series data, weighted filtering processing may be performed on the characteristic operation benchmark time series data.

[0064] For each sample point x[n] of the signal, its noise-reduced value y[n] can be calculated by the following formula:

[0065]

[0066] Where w k is the weight of the offset k relative to the window center point n, M is the number of weights, and x[n+k] is the corresponding feature operation benchmark time series data in the window.

[0067] Step S3: Perform warping on the multiple characteristic operation benchmark time series data based on the dynamic time warping method to determine the optimal warping path benchmark curve of the multiple characteristic operation benchmark time series data.

[0068] In an embodiment of the present invention, the dynamic time warping method introduces the idea of ​​sparsity. Starting from the perspective of the time series itself, the search range of the matrix cells is set according to the difference between the two sequences. This can ensure that on the basis of the accuracy of the final calculation result obtained by the algorithm, a certain amount of space complexity is reduced, so as to achieve the effect of reducing calculation time and improving accuracy.

[0069] Specifically, step S3 includes:

[0070] Step S301: construct an n×m sparse matrix of multiple feature operation benchmark time series data, where n represents the number of columns of cells in the sparse matrix, and m represents the number of rows of cells in the sparse matrix.

[0071] Among them, the lengths of multiple feature operation benchmark time series data are the same, n is the number of sampling points, and m is the length of the feature operation benchmark time series data.

[0072] Step S302: Initialize the cells in the sparse matrix to null values.

[0073] Among them, the sparse matrix is ​​converted into a matrix form consisting of multiple cells, such as Figure 2 The cells of the sparse matrix at the time of initialization are shown, wherein each cell is empty, and the position of the empty value is not within the range of the final search path.

[0074] Step S303: normalize the multiple feature operation benchmark time series data.

[0075] Normalization is the ratio of the difference between each point in the sequence and the minimum value, and the difference between the maximum and minimum values. The goal is to use a formula to set the values ​​of all points in the two sequences to the range [0, 1], facilitating subsequent classification and comparison.

[0076] Step S304: Calculate the mean of all normalized feature operation benchmark time series data to obtain a total mean.

[0077] Among them, the total mean is determined based on the ratio of the sum of all normalized feature operation benchmark time series data to the number of all normalized feature operation benchmark time series data. In some embodiments, the standard deviation, variance, etc. can also be calculated to replace the total mean.

[0078] Step S305: perform difference operations on the multiple normalized characteristic operation benchmark time series data and the total mean respectively, fill in the corresponding cells in the sparse matrix according to the differences, and form a search path range.

[0079] It should be noted that when running benchmark time series data with multiple features in real applications, the data length is very long. Calculating the initial and cumulative distances requires calculating the value of each cell in the matrix, which results in high computational complexity and large space requirements. However, the dynamic time warping algorithm, which incorporates sparsity, reduces computational complexity by optimizing the difference between the benchmark time series data and the overall mean, thereby reducing time complexity.

[0080] Specifically, step S305 includes:

[0081] Step S3051: perform difference calculations on the multiple normalized characteristic operation benchmark time series data and the total mean respectively, and determine whether the difference is less than a preset difference threshold.

[0082] Step S3052: If it is determined that the difference is less than a preset difference threshold, the element of the cell corresponding to the normalized characteristic operation benchmark time series data in the sparse matrix is ​​updated to the difference.

[0083] Step S3053: If it is determined that the difference is not less than the preset difference threshold, the element of the cell corresponding to the normalized characteristic operation benchmark time series data in the sparse matrix is ​​still a null value.

[0084] The difference threshold may be set based on experience, preferably 0.25, 0.3 or 0.35.

[0085] For example, the difference threshold is set to 0.25 and the total mean is set to 0.35. If the two normalized feature operation benchmark time series data are 0.7 and 0.32 respectively, then the differences between the two normalized feature operation benchmark time series data and the total mean are 0.35 and 0.03 respectively (the differences are all absolute values). Then, the elements of the cells corresponding to the normalized feature operation benchmark time series data 0.32 in the sparse matrix need to be updated to 0.03, while the elements of the cells corresponding to the normalized feature operation benchmark time series data 0.7 in the sparse matrix are still empty values.

[0086] Step S3054: Filter out cells corresponding to non-null values ​​of the sparse matrix to form an initial search path range.

[0087] like Figure 3 As shown, Figure 3 The initial search path range of the sparse matrix is ​​illustrated, where the initial search path range is gray.

[0088] Step S3055: Determine whether there is an isolated cell in the sparse matrix. The principle for determining an isolated cell is that all adjacent cells of the cell are null values.

[0089] It should be noted that since the search path needs to be traversed continuously and a continuous search path is formed, and since the search path is relatively sparse, null values ​​may be encountered in the final path traversal. In order to ensure that a continuous backtracking path is obtained in the end, we still need to fill in some isolated cells. However, these isolated cells are relatively few, so they have little impact on the subsequent curve fitting. Figure 3 As shown, Figure 3 The dark black cells in the image are isolated cells.

[0090] Step S3056: If it is determined that there are isolated cells in the sparse matrix, the adjacent cells corresponding to the isolated cells are filled with values, and the values ​​filled in the adjacent cells are the differences between the normalized feature operation benchmark time series data corresponding to the cells and the total mean.

[0091] If there is no isolated cell in the sparse matrix, continue to execute step S3047.

[0092] Step S3057: Update the initial search path range according to the filled cells to form a search path range.

[0093] Among them, the filled cells are added to the initial search path range to form the final search path range.

[0094] Step S306: Within the search path range, starting from the cells with non-empty values ​​in the bottom row of the sparse matrix, traverse upward in sequence until reaching the non-empty values ​​in the top row of the sparse matrix, and obtain multiple search paths and the cumulative distance values ​​corresponding to the multiple search paths.

[0095] The starting point is determined as follows: starting from the cell with a non-empty value in the bottom row of the sparse matrix, it is marked as the current position.

[0096] Search upward: Starting from the current position, search the three adjacent cells (upper left, upper, left) in the upward row, find the cell with the smallest value, and mark it as the next current position. In each traversal, only one current position is marked in each row.

[0097] End point determination: Repeat the upward search steps and continuously update the current position until the non-empty value of the topmost row in the sparse matrix is ​​reached, which is the end point of the search path.

[0098] Path recording: During the search process, the cells passed through are recorded. These cells constitute the search path. Since there are multiple non-null values ​​in the bottom row of the sparse matrix, and multiple different branches are likely to appear during the search process, multiple search paths can be formed.

[0099] The process of calculating the cumulative distance value corresponding to the search path includes:

[0100] Step S3061: Calculate the Euclidean distance between every two adjacent cells according to the positions of every two adjacent cells in the search path.

[0101] Step S3062: Accumulate all Euclidean distances in the search path to obtain a cumulative distance value of the search path.

[0102] Step S307 : Filter out the search path with the smallest cumulative distance value, perform curve fitting on the filtered search path, and obtain the optimal regularized path reference curve of the multiple characteristic operation reference time series data.

[0103] For example, when the difference threshold is set to 0.25, Figure 4 As shown, Figure 4 The optimal regularized path reference curve is shown.

[0104] Step S4: numerically fitting the characteristic operation sequence data of the current operation period of the circuit breaker to be tested to obtain a characteristic operation sequence path curve of the current operation period of the circuit breaker to be tested.

[0105] The data types of the characteristic operation sequence data of the current operation period of the circuit breaker to be tested and the characteristic operation reference sequence data are consistent, and the characteristic operation sequence path curve is obtained by performing numerical curve fitting on the characteristic operation sequence data.

[0106] Step S5: comparing the similarity between the characteristic operation sequence path curve of the current operation period of the circuit breaker to be tested and the optimal regularized path reference curve.

[0107] The similarity can be determined by calculating the Euclidean distance between the characteristic operation sequence path curve of the current operation period of the circuit breaker to be tested and the optimal regularized path reference curve.

[0108] It can be understood that if the similarity is high, it means that the characteristic operation sequence path curve of the current operation period of the circuit breaker to be tested fits the optimal regular path reference curve, and the circuit breaker to be tested is determined to be in a normal state; otherwise, the circuit breaker to be tested is determined to be in a fault state.

[0109] Step S6: When the similarity is less than the preset similarity threshold, it is determined that the current operating period of the circuit breaker to be tested is in a fault state.

[0110] It should be noted that the present invention uses a dynamic time warping method to warp multiple characteristic operating benchmark time series data of the circuit breaker under normal operation, and obtains the optimal warping path benchmark curve about the time series, so as to intuitively characterize the operating trend of the circuit breaker under normal operation, and judge the degree of similarity between the characteristic operating sequence path curve of the current operating period of the circuit breaker to be tested and the optimal warping path benchmark curve, so as to judge the fault state of the circuit breaker to be tested, so that the fault state of the circuit breaker to be tested can be estimated in real time, and the accuracy of the fault state assessment of the circuit breaker is improved.

[0111] The above is a detailed description of an embodiment of a circuit breaker mechanical fault analysis method based on dynamic time warping provided by the present invention. The following is a detailed description of an embodiment of a circuit breaker mechanical fault analysis system based on dynamic time warping provided by the present invention.

[0112] like Figure 5 As shown, the present invention also provides a circuit breaker mechanical fault analysis system based on dynamic time warping method, comprising:

[0113] The reference data acquisition module 100 is used to acquire a plurality of characteristic operation reference time series data of the circuit breaker under normal operation, wherein the data types of the plurality of characteristic operation reference time series data are the same;

[0114] A preprocessing module 200 is used to preprocess multiple feature operation benchmark time series data;

[0115] A time warping module 300 is configured to warp a plurality of characteristic operation benchmark time series data based on a dynamic time warping method, and determine an optimal warping path reference curve for the plurality of characteristic operation benchmark time series data;

[0116] The numerical fitting module 400 is used to perform numerical fitting on the characteristic operation sequence data of the current operation period of the circuit breaker to be tested, so as to obtain a characteristic operation sequence path curve of the current operation period of the circuit breaker to be tested;

[0117] A similarity module 500 is used to compare the similarity between the characteristic operation sequence path curve of the current operation period of the circuit breaker to be tested and the optimal regularized path reference curve;

[0118] The fault judgment module 600 is configured to determine that the current operating period of the circuit breaker to be tested is in a fault state when the similarity is less than a preset similarity threshold.

[0119] like Figure 6 As shown, the present invention also provides an electronic device 10, including a memory 20 and a processor 30, wherein a computer program is stored in the memory 20. When the computer program is executed by the processor 30, the processor 30 executes the steps of the circuit breaker mechanical fault analysis method based on the dynamic time warping method as described above.

[0120] The present invention also provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of the circuit breaker mechanical fault analysis method based on the dynamic time warping method as described above when the computer program is executed.

[0121] The present invention also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions, wherein when the program instructions are executed by the computer, the computer executes the steps of the circuit breaker mechanical fault analysis method based on the dynamic time warping method as described above.

[0122] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, electronic devices, computer storage media, and computer program products can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0123] In several embodiments provided by the present invention, it is understood that each box in the flow chart or block diagram can represent a module, program segment or part of the code, and the module, program segment or part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order than that marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved.

[0124] In the several embodiments provided by the present invention, it should be understood that the disclosed systems, electronic devices, computer storage media, computer program products and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0125] Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0126] In addition, the functional units in the various embodiments of the present invention may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0127] If the integrated unit is implemented in the form of 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, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for executing all or part of the steps of the method described in each embodiment of the present invention through a computer device (which can be a personal computer, server, or network device, etc.). The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (full name: Read-Only Memory, English abbreviation: ROM), random access memory (full name: Random Access Memory, English abbreviation: RAM), magnetic disk or optical disk, etc. Various media that can store program codes.

[0128] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A circuit breaker mechanical fault analysis method based on dynamic time warping method, characterized in that: include: Acquire multiple characteristic operation benchmark time series data of the circuit breaker under normal operation, where the data types of the multiple characteristic operation benchmark time series data are the same; Preprocessing a plurality of characteristic operation benchmark time series data; Performing warping on the plurality of characteristic operation benchmark time series data based on a dynamic time warping method to determine an optimal warping path benchmark curve for the plurality of characteristic operation benchmark time series data includes: Constructing multiple n×m sparse matrices of the characteristic operation benchmark time series data, where n represents the number of columns of cells in the sparse matrix, and m represents the number of rows of cells in the sparse matrix; Initializing cells in the sparse matrix to empty values; Normalizing the plurality of characteristic operation benchmark time series data; Calculate the mean of all normalized feature-based benchmark time series data to obtain a total mean; Performing difference operations on multiple normalized feature operation benchmark time series data and the total mean respectively, filling values ​​in corresponding cells in the sparse matrix according to the differences, and forming a search path range, including: Performing difference calculations on the plurality of normalized characteristic operation benchmark time series data and the total mean value, and determining whether the difference is less than a preset difference threshold; If it is determined that the difference is less than the preset difference threshold, updating the element of the cell corresponding to the normalized feature operation benchmark time series data in the sparse matrix to the difference; If it is determined that the difference is not less than the preset difference threshold, the element of the cell corresponding to the normalized feature operation benchmark time series data in the sparse matrix remains a null value; Filter out cells corresponding to non-empty values ​​of the sparse matrix to form an initial search path range; Determine whether there is an isolated cell in the sparse matrix, and the principle for determining the isolated cell is: the adjacent cells of the cell are all null values; If it is determined that the isolated cell exists in the sparse matrix, the adjacent cells corresponding to the isolated cell are filled with values, and the values ​​filled in the adjacent cells are the differences between the normalized feature operation benchmark time series data corresponding to the cell and the total mean; Update the initial search path range according to the filled cells to form a search path range; Within the search path range, starting from the cells with non-empty values ​​in the bottom row of the sparse matrix, traverse upward in sequence until reaching the cells with non-empty values ​​in the top row of the sparse matrix, to obtain multiple search paths and cumulative distance values ​​corresponding to the multiple search paths; Screening out a search path with the smallest cumulative distance value, performing curve fitting on the screened search path, and obtaining an optimal regularized path reference curve for a plurality of characteristic operation reference time series data; Performing numerical fitting on characteristic operation timing data of the current operation period of the circuit breaker to be tested to obtain a characteristic operation timing path curve of the current operation period of the circuit breaker to be tested; Comparing the similarity between the characteristic operation sequence path curve of the current operation period of the circuit breaker to be tested and the optimal regularized path reference curve; When the similarity is less than a preset similarity threshold, it is determined that the current operating period of the circuit breaker to be tested is in a fault state.

2. The circuit breaker mechanical failure analysis method based on dynamic time warping method according to claim 1, characterized in that: The data type of the characteristic operation reference timing data is opening and closing stroke timing data or opening and closing coil current timing data.

3. The circuit breaker mechanical failure analysis method based on dynamic time warping method according to claim 1, characterized in that: The pre-processing includes weighted filtering.

4. The circuit breaker mechanical failure analysis method based on dynamic time warping method according to claim 1, characterized in that: Also includes: Calculating the Euclidean distance between each two adjacent cells according to the positions of each two adjacent cells in the search path; All Euclidean distances in the search path are accumulated to obtain a cumulative distance value of the search path.

5. A circuit breaker mechanical fault analysis system based on dynamic time warping method, characterized in that: include: A reference data acquisition module is used to acquire a plurality of characteristic operation reference time series data of the circuit breaker under normal operation, wherein the data types of the plurality of characteristic operation reference time series data are the same; A preprocessing module, configured to preprocess the plurality of characteristic operation benchmark time series data; A time series warping module is used to warp the plurality of characteristic operation benchmark time series data based on a dynamic time warping method, and determine an optimal warping path reference curve for the plurality of characteristic operation benchmark time series data; Performing warping on the plurality of characteristic operation benchmark time series data based on a dynamic time warping method to determine an optimal warping path benchmark curve for the plurality of characteristic operation benchmark time series data includes: Constructing multiple n×m sparse matrices of the characteristic operation benchmark time series data, where n represents the number of columns of cells in the sparse matrix, and m represents the number of rows of cells in the sparse matrix; Initializing cells in the sparse matrix to empty values; Normalizing the plurality of characteristic operation benchmark time series data; Calculate the mean of all normalized feature-based benchmark time series data to obtain a total mean; Performing difference operations on multiple normalized feature operation benchmark time series data and the total mean respectively, filling values ​​in corresponding cells in the sparse matrix according to the differences, and forming a search path range, including: Performing difference calculations on the plurality of normalized characteristic operation benchmark time series data and the total mean value, and determining whether the difference is less than a preset difference threshold; If it is determined that the difference is less than the preset difference threshold, updating the element of the cell corresponding to the normalized feature operation benchmark time series data in the sparse matrix to the difference; If it is determined that the difference is not less than the preset difference threshold, the element of the cell corresponding to the normalized feature operation benchmark time series data in the sparse matrix remains a null value; Filter out cells corresponding to non-empty values ​​of the sparse matrix to form an initial search path range; Determine whether there is an isolated cell in the sparse matrix, and the principle for determining the isolated cell is: the adjacent cells of the cell are all null values; If it is determined that the isolated cell exists in the sparse matrix, the adjacent cells corresponding to the isolated cell are filled with values, and the values ​​filled in the adjacent cells are the differences between the normalized feature operation benchmark time series data corresponding to the cell and the total mean; Update the initial search path range according to the filled cells to form a search path range; Within the search path range, starting from the cells with non-empty values ​​in the bottom row of the sparse matrix, traverse upward in sequence until reaching the cells with non-empty values ​​in the top row of the sparse matrix, to obtain multiple search paths and cumulative distance values ​​corresponding to the multiple search paths; Screening out a search path with the smallest cumulative distance value, performing curve fitting on the screened search path, and obtaining an optimal regularized path reference curve for a plurality of characteristic operation reference time series data; A numerical fitting module is used to perform numerical fitting on characteristic operation sequence data of the current operation period of the circuit breaker to be tested, so as to obtain a characteristic operation sequence path curve of the current operation period of the circuit breaker to be tested; A similarity module, configured to compare the similarity between the characteristic operation sequence path curve of the current operation period of the circuit breaker to be tested and the optimal regularized path reference curve; The fault judgment module is configured to judge that the current operation period of the circuit breaker to be tested is in a fault state when the similarity is less than a preset similarity threshold.

6. An electronic device, characterized in that: The method comprises a memory and a processor, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, the processor executes the steps of the circuit breaker mechanical fault analysis method based on the dynamic time warping method as described in any one of claims 1 to 4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed, the steps of the circuit breaker mechanical failure analysis method based on the dynamic time warping method according to any one of claims 1 to 4 are implemented.

8. A computer program product, characterized in that The computer program product includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions, wherein when the program instructions are executed by a computer, the computer is caused to perform the steps of the circuit breaker mechanical failure analysis method based on the dynamic time warping method as described in any one of claims 1 to 4.

Images