A method for analyzing contact resistance of a cable intermediate joint and related apparatus

Abstract

The application discloses a contact resistance analysis method and related device of cable intermediate joint, which decomposes original time domain signal into two-dimensional time-frequency information by using discrete wavelet transform (DWT), improves the performance of local signal, and improves the anti-noise performance of subsequent prediction model, and still has good anti-noise ability for data with large fluctuation range; a self-suggested support vector machine model (SSVM) extracts subsequent knowledge from a training stage by using a non-iterative method, i.e., knowledge extracted from the second type of support vector, avoids obtaining information from the training stage as the first type of support vector or the position of the hyperplane in the test stage, thereby avoiding false classification and false learning, and improving the prediction accuracy of the contact resistance. Thus, the technical problem of low contact resistance prediction accuracy of the cable intermediate joint in the prior art is solved.

Description

Technical Field

[0001] This application relates to the field of power cable technology, and in particular to a method and related apparatus for analyzing the contact resistance of cable intermediate joints. Background Technology

[0002] Cables consist of two parts: the cable body and the cable joint. Compared to the cable body, the cable joint has a more complex structure and involves more uncertainties during on-site construction. For the entire cable, the weakest point, the intermediate joint, affects its current-carrying capacity, and contact resistance is a key factor influencing the temperature rise of the intermediate joint. While the structure and materials of the intermediate joint are complex, they can be calculated using finite element simulation models. However, due to on-site construction, the contact resistance at the joint cannot be accurately calculated, and it changes over time due to factors such as cable material changes and insulation lifespan degradation. Therefore, it is necessary to study a non-invasive contact resistance inversion method to adapt to the actual engineering needs of cables in operation, accurately grasp the magnitude of the contact resistance, and promptly understand the cable temperature rise characteristics for fault prevention and early warning.

[0003] Inversion methods refer to artificial intelligence systems that can mimic human intelligence, possessing learning and reasoning capabilities. Common inversion methods include time series forecasting, recurrent neural networks (RNNs), long short-term memory artificial neural network models (LSTM), and support vector machines (SVMs). Time series analysis is not adept at handling nonlinear data. RNNs are suitable for time series data forecasting, but suffer from problems such as vanishing gradients, exploding gradients, and poor information dependence on long-range sequence data. LSTM models can address the aforementioned problems of RNNs, but for data with large fluctuations, a single LSTM model cannot guarantee prediction accuracy. Traditional SVMs ignore training data that is not linearly separated by the kernel during the training phase. If the data is similar to or identical to misclassified data in the test set, it will be misclassified again because data close to the misclassified data is uncertain. Summary of the Invention

[0004] This application provides a method and related apparatus for analyzing the contact resistance of cable intermediate joints, which solves the technical problem of low accuracy in predicting the contact resistance of cable intermediate joints in the prior art.

[0005] In view of this, the first aspect of this application provides a method for analyzing the contact resistance of a cable intermediate joint, the method comprising:

[0006] Obtain the original time-domain signal of the cable corresponding to the intermediate joint of the cable to be analyzed;

[0007] The original time-domain signal is discretized into low-frequency and high-frequency signals by discrete wavelet transform, and then the low-frequency and high-frequency signals are reconstructed by fundamental wavelet filter, low-pass filter and high-pass filter to obtain the reconstructed signal;

[0008] A self-advancement support vector machine is constructed. The reconstructed signal is input into the self-advancement support vector machine for training. The misclassified data of the reconstructed signal is obtained and weights are assigned. The misclassified data of the reconstructed signal is removed according to the weights to obtain the final signal.

[0009] The final signal is input into the self-suggested support vector machine for prediction to obtain the contact resistance of the intermediate joint of the cable to be analyzed.

[0010] Optionally, the step of discretizing the original time-domain signal into low-frequency and high-frequency signals using discrete wavelet transform, and then reconstructing the low-frequency and high-frequency signals using a fundamental wavelet filter, a low-pass filter, and a high-pass filter to obtain the reconstructed signal, specifically includes:

[0011] The original time-domain signal is discretized into low-frequency and high-frequency signals using discrete wavelet transform.

[0012] The low-frequency signal and the high-frequency signal are input into a preset reconstruction formula to obtain the reconstructed signal;

[0013] The preset reconstruction formula is as follows:

[0014]

[0015] In the formula, c is a constant, and A n For low-frequency signals, D i Let f(t) be a high-frequency signal, and l(ψ) be the reconstructed signal obtained from the inversion. jk (t) represents a low-pass filter, h(ψ) jk (t) represents a high-pass filter.

[0016] Optionally, the step of constructing a self-advancement support vector machine involves inputting the reconstructed signal into the self-advancement support vector machine for training, obtaining misclassified data of the reconstructed signal and assigning weights to it, and then removing the misclassified data of the reconstructed signal according to the weights to obtain the final signal. Specifically, this includes:

[0017] Construct a self-advancement support vector machine, and input the reconstructed signal into the self-advancement support vector machine for training;

[0018] when If so, then the weight of the misclassified data is obtained according to the preset weight calculation formula, and if the calculated weight is zero, x is determined to be misclassified. i This is non-misclassified data;

[0019] when The weights of the misclassified data are obtained through a preset weighting formula, and the misclassified data in the reconstructed signal are removed according to the weights to obtain the final signal.

[0020] in:

[0021] In the formula, MD is the misclassified dataset, and α j ≥0, b and k are constants; x i x j y i y j j = 1, ..., N are training data that do not belong to the MD set;

[0022] The preset weight calculation formula is as follows:

[0023]

[0024] In the formula, AW is the weight, and h(x) is the single original data.

[0025] Optionally, the step of inputting the final signal into the self-proposing support vector machine for prediction to obtain the contact resistance of the intermediate joint of the cable to be analyzed further includes:

[0026] The first contact resistance of the cable joint to be analyzed was calculated using conventional calculation methods, and the second contact resistance of the cable joint was measured using a micro-ohmmeter.

[0027] An analysis comparison table is generated based on the first contact resistance, the second contact resistance, and the contact resistance of the intermediate joint of the cable to be analyzed.

[0028] A second aspect of this application provides a contact resistance analysis system for cable intermediate joints, the system comprising:

[0029] The acquisition module is used to acquire the original time-domain signal of the cable corresponding to the intermediate joint of the cable to be analyzed;

[0030] The reconstruction module is used to discretize the original time-domain signal into low-frequency and high-frequency signals through discrete wavelet transform, and then reconstruct the low-frequency and high-frequency signals through a fundamental wavelet filter, a low-pass filter, and a high-pass filter to obtain the reconstructed signal.

[0031] The filtering module is used to construct a self-advancement support vector machine. The reconstructed signal is input into the self-advancement support vector machine for training, the misclassified data of the reconstructed signal is obtained and weights are assigned. The misclassified data of the reconstructed signal is removed according to the weights to obtain the final signal.

[0032] The analysis module is used to input the final signal into the self-suggested support vector machine for prediction to obtain the contact resistance of the intermediate joint of the cable to be analyzed.

[0033] Optionally, the reconstruction module is specifically used for:

[0034] The original time-domain signal is discretized into low-frequency and high-frequency signals using discrete wavelet transform.

[0035] The low-frequency signal and the high-frequency signal are input into a preset reconstruction formula to obtain the reconstructed signal;

[0036] The preset reconstruction formula is as follows:

[0037]

[0038] In the formula, c is a constant, and A n For low-frequency signals, D i Let f(t) be a high-frequency signal, and l(ψ) be the reconstructed signal obtained from the inversion. jk (t) represents a low-pass filter, h(ψ) jk (t) represents a high-pass filter.

[0039] Optionally, the filtering module is specifically used for:

[0040] Construct a self-advancement support vector machine, and input the reconstructed signal into the self-advancement support vector machine for training;

[0041] when If so, then the weight of the misclassified data is obtained according to the preset weight calculation formula, and if the calculated weight is zero, x is determined to be misclassified. i This is non-misclassified data;

[0042] when The weights of the misclassified data are obtained through a preset weighting formula, and the misclassified data in the reconstructed signal are removed according to the weights to obtain the final signal.

[0043] in:

[0044]

[0045] In the formula, MD is the misclassified dataset, and α j ≥0, b and k are constants; x i x j y i y j j = 1, ..., N are training data that do not belong to the MD set;

[0046] The preset weight calculation formula is as follows:

[0047]

[0048] In the formula, AW is the weight, and h(x) is the single original data.

[0049] Optionally, it also includes: a generation module;

[0050] The generation module is used to calculate the first contact resistance of the intermediate joint of the cable to be analyzed using conventional calculation methods, and to measure the second contact resistance of the intermediate joint of the cable using a micro-ohmmeter; and to generate an analysis comparison table based on the first contact resistance, the second contact resistance and the contact resistance of the intermediate joint of the cable to be analyzed.

[0051] A third aspect of this application provides a contact resistance analysis device for cable joints, the device comprising a processor and a memory:

[0052] The memory is used to store program code and transmit the program code to the processor;

[0053] The processor is configured to execute the steps of the contact resistance analysis method for cable intermediate joints as described in the first aspect above, according to the instructions in the program code.

[0054] A fourth aspect of this application provides a computer-readable storage medium for storing program code for performing the method described in the first aspect above.

[0055] As can be seen from the above technical solutions, this application has the following advantages:

[0056] This application provides a method for analyzing the contact resistance of a cable joint, comprising: acquiring the original time-domain signal of the cable corresponding to the cable joint to be analyzed; discretizing the original time-domain signal into low-frequency and high-frequency signals through discrete wavelet transform, and then reconstructing the low-frequency and high-frequency signals through a fundamental wavelet filter, a low-pass filter, and a high-pass filter to obtain a reconstructed signal; constructing a self-advising support vector machine, inputting the reconstructed signal into the self-advising support vector machine for training, obtaining misclassified data of the reconstructed signal and assigning weights, and removing the misclassified data of the reconstructed signal according to the weights to obtain the final signal; and inputting the final signal into the self-advising support vector machine for prediction to obtain the contact resistance of the cable joint to be analyzed.

[0057] Compared with the prior art, this application:

[0058] (1) The original time-domain signal is decomposed into two-dimensional time-frequency information by using Discrete Wavelet Transform (DWT), which improves the performance of local signals and enhances the noise resistance of subsequent prediction models. It still has good noise resistance for data with large fluctuation range.

[0059] (2) The self-suggested support vector machine model (SSVM) uses a non-iterative method to extract subsequent knowledge from the training phase, that is, knowledge extracted from the second type of support vectors. This avoids the information obtained from the training phase in the testing phase being the first type of support vectors or the hyperplane position. It considers the proximity between the test data and the misclassified data, reflects and corrects the original misclassified data, thereby avoiding misclassified data and improving the prediction accuracy of contact resistance. Attached Figure Description

[0060] Figure 1 This is a flowchart illustrating an embodiment of a method for analyzing the contact resistance of a cable intermediate joint provided in this application.

[0061] Figure 2 This is a schematic diagram of a contact resistance analysis system for a cable intermediate joint provided in this application. Detailed Implementation

[0062] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present application.

[0063] Please see Figure 1 The method for analyzing the contact resistance of a cable intermediate joint provided in this application includes:

[0064] Step 101: Obtain the original time-domain signal of the cable corresponding to the intermediate joint of the cable to be analyzed;

[0065] Step 102: Discretize the original time-domain signal into low-frequency and high-frequency signals using discrete wavelet transform, and then reconstruct the low-frequency and high-frequency signals using a fundamental wavelet filter, a low-pass filter, and a high-pass filter to obtain the reconstructed signal.

[0066] It should be noted that:

[0067] 1) Discrete wavelet transform

[0068] Wavelet transform is divided into two types: continuous wavelet transform (CWT) and discrete wavelet transform (DWT). CWT basis wavelet ψ ab The transformation formula and CWT formula are as follows:

[0069]

[0070]

[0071] Where a is the scale factor (a>0) and b is the shift factor (b∈R). Adjusting a and b allows the time series signal to be decomposed into high-frequency and low-frequency time coefficients, WT f (a,b) represents the continuous wavelet coefficients, and f(t) represents the original data. and ψ ab (t) are conjugate functions.

[0072] DWT is obtained by discretizing the continuous wavelet transform with powers of 2 in terms of scale and displacement, and its basis wavelet ψ jk The transformation formula and DWT formula are as follows:

[0073]

[0074]

[0075]

[0076] Where a0 > 0, b0 ∈ R, k is a non-negative integer, ψ mn (t) represents the basis wavelet of the DWT wavelet transform, WT f (m,n) represent the discrete wavelet coefficients, and f(t) represents the original data. and ψ mn (t) are conjugate functions.

[0077] This application discretizes the original time-domain signal into low-frequency and high-frequency signals using discrete wavelet transform, and then obtains the low-frequency signal A by reconstructing the high- and low-frequency wavelet coefficients. n and high-frequency signal D n The former can represent general information, while the latter can represent detailed information. Adding the two together can reconstruct the original information, and the reconstruction formula is as follows:

[0078]

[0079] Where c is a constant, A n For low-frequency signals, D i Let f(t) be a high-frequency signal, and l(ψ) be the reconstructed signal obtained from the inversion. jk (t) represents a low-pass filter, h(ψ) jk (t) represents a high-pass filter.

[0080] Step 103: Construct a self-advancement support vector machine. Input the reconstructed signal into the self-advancement support vector machine for training. Obtain the misclassified data of the reconstructed signal and assign weights. Remove the misclassified data of the reconstructed signal according to the weights to obtain the final signal.

[0081] It should be noted that:

[0082] 2) Self-advised support vector machine prediction

[0083] Traditional SVM ignores training data that is not linearly distributed by the kernel during training, potentially leading to misclassification. Therefore, the self-proposing support vector machine prediction method proposed in this invention considers the similarity between test data and misclassified data, thereby reflecting and correcting the original misclassified data.

[0084] The misclassified dataset MD is as follows:

[0085]

[0086] In the formula, α j ≥0, b and k are constants.

[0087] Ideally, the neighborhood set (MD) is empty, but experimental results show that misclassified data is a common phenomenon. For x in the MD, the neighborhood length NL is:

[0088] NL(x i ) = min imum xj (x i -x j |y i ≠y j ) / 2 (8)

[0089] In the formula, x i x j y i y j j = 1, ..., N are training data that do not belong to the MD set.

[0090] Based on the results of NL, x i The weights can represent how close the test data is to the misclassified data, as shown below:

[0091]

[0092] In the formula, AW is the weight, and h(x) is the single original data.

[0093] Specifically, a self-advancement support vector machine is constructed, and the reconstructed signal is input into the self-advancement support vector machine for training;

[0094] when If so, then the weight of the misclassified data is obtained according to the preset weight calculation formula, and if the calculated weight is zero, x is determined to be misclassified. i This is non-misclassified data;

[0095] when The weights of misclassified data are obtained by using a preset weight calculation formula, and the misclassified data of the reconstructed signal are removed according to the weights to obtain the final signal.

[0096] Step 104: Input the final signal into the self-proposed support vector machine for prediction to obtain the contact resistance of the intermediate joint of the cable to be analyzed.

[0097] Furthermore, in one embodiment, the first contact resistance of the cable joint to be analyzed is calculated using conventional calculation methods, and the second contact resistance of the cable joint is measured using a micro-ohmmeter; an analysis comparison table is generated based on the first contact resistance, the second contact resistance, and the contact resistance of the cable joint to be analyzed.

[0098] Specifically, by using the first 90% of the test data obtained by applying different currents to single-core cables as the sample set and the last 10% of the experimental data as the validation set, some prediction results of the method proposed in this invention are shown in the table below. The method proposed in this invention has high prediction accuracy and fast calculation speed.

[0099] Table 1 Comparison of Prediction Results

[0100] Measured resistance value (μΩ) Conventional resistance calculation method (μΩ) The calculation method of this invention (μΩ) 5.1μΩ R = 6.3 μΩ 5.3μΩ

[0101] The above is a method for analyzing the contact resistance of a cable intermediate joint provided in the embodiments of this application. The following is a system for analyzing the contact resistance of a cable intermediate joint provided in the embodiments of this application.

[0102] Please see Figure 2 The contact resistance analysis system for cable intermediate joints provided in this application embodiment includes:

[0103] The acquisition module 201 is used to acquire the original time-domain signal of the cable corresponding to the intermediate joint of the cable to be analyzed;

[0104] The reconstruction module 202 is used to discretize the original time-domain signal into low-frequency and high-frequency signals through discrete wavelet transform, and then reconstruct the low-frequency and high-frequency signals through a fundamental wavelet filter, a low-pass filter, and a high-pass filter to obtain the reconstructed signal.

[0105] The filtering module 203 is used to construct a self-advancement support vector machine. The reconstructed signal is input into the self-advancement support vector machine for training, the misclassified data of the reconstructed signal is obtained and weights are assigned. The misclassified data of the reconstructed signal is removed according to the weights to obtain the final signal.

[0106] Analysis module 204 is used to input the final signal into a self-proposed support vector machine for prediction to obtain the contact resistance of the intermediate joint of the cable to be analyzed.

[0107] Furthermore, this application embodiment also provides a contact resistance analysis device for cable intermediate joints, the device including a processor and a memory:

[0108] The memory is used to store program code and transmit the program code to the processor;

[0109] The processor is used to execute the steps of the contact resistance analysis method for cable intermediate joints as described in the above method embodiments, according to the instructions in the program code.

[0110] Furthermore, this application embodiment also provides a computer-readable storage medium for storing program code for executing the contact resistance analysis method for cable intermediate joints described in the above method embodiment.

[0111] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process of the system and unit described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0112] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented, for example, in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0113] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0114] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection between apparatuses or units through some interfaces, and may be electrical, mechanical, or other forms.

[0115] The units described as separate components may or may not be physically separate. The 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 the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0116] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0117] If the integrated unit 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 this application, in essence, 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. 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 this application. 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.

[0118] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method of analyzing contact resistance of a cable intermediate joint, characterized by, The method comprises the following steps: acquiring an original time domain signal of a cable corresponding to a cable intermediate joint to be analyzed; discretizing the original time domain signal into a low-frequency signal and a high-frequency signal through discrete wavelet transform, and reconstructing the low-frequency signal and the high-frequency signal through a base wavelet filter, a low-pass filter and a high-pass filter to obtain a reconstructed signal; constructing a self-advisory support vector machine, inputting the reconstructed signal into the self-advisory support vector machine for training, obtaining error classification data of the reconstructed signal and assigning a weight, and removing the error classification data of the reconstructed signal according to the weight to obtain a final signal; inputting the final signal into the self-advisory support vector machine for prediction to obtain a contact resistance of the cable intermediate joint to be analyzed; wherein the step of constructing the self-advisory support vector machine, inputting the reconstructed signal into the self-advisory support vector machine for training, obtaining error classification data of the reconstructed signal and assigning a weight, and removing the error classification data of the reconstructed signal according to the weight to obtain a final signal specifically comprises: constructing a self-advisory support vector machine, inputting the reconstructed signal into the self-advisory support vector machine for training; when If so, the weight of the misclassified data is obtained according to the preset weight calculation formula, and if the calculated weight is zero, then a decision is made. This is non-misclassified data; When , the weight of the misclassified data is obtained by a preset weight calculation formula, and the misclassified data of the reconstructed signal is removed according to the weight to obtain a final signal; wherein: ； ； In the formula, For misclassified datasets, , and It is a constant; , , , j=1, ..., N, are not belonging to Centralized training data; the preset weight calculation formula is: ； In the formula, is a weight, is a single raw data.

2. The contact resistance analysis method of a cable intermediate joint according to claim 1, characterized by, the step of discretizing the original time domain signal into a low-frequency signal and a high-frequency signal through discrete wavelet transform, and reconstructing the low-frequency signal and the high-frequency signal through a base wavelet filter, a low-pass filter and a high-pass filter to obtain a reconstructed signal specifically comprises: discretizing the original time domain signal into a low-frequency signal and a high-frequency signal through discrete wavelet transform; inputting the low-frequency signal and the high-frequency signal into a preset reconstruction formula to obtain the reconstructed signal; wherein the preset reconstruction formula is: ； wherein is a constant, is a low frequency signal, is a high frequency signal, denotes the reconstructed signal of the inversion, denotes a low pass filter, denotes a high pass filter.

3. The contact resistance analysis method of a cable intermediate joint according to claim 1, characterized by, the step of inputting the final signal into the self-advisory support vector machine for prediction to obtain a contact resistance of the cable intermediate joint to be analyzed further comprises: calculating a first contact resistance of the cable intermediate joint to be analyzed through a conventional calculation method, and measuring a second contact resistance of the cable intermediate joint through a micro-ohm meter; generating an analysis comparison table according to the first contact resistance, the second contact resistance and the contact resistance of the cable intermediate joint to be analyzed.

4. A contact resistance analysis system for cable intermediate joints, characterized by The method comprises the following steps: an acquisition module is configured to acquire an original time domain signal of a cable corresponding to a cable intermediate joint to be analyzed; a reconstruction module is configured to discretize the original time domain signal into a low-frequency signal and a high-frequency signal through discrete wavelet transform, and reconstruct the low-frequency signal and the high-frequency signal through a base wavelet filter, a low-pass filter and a high-pass filter to obtain a reconstructed signal; a screening module is configured to construct a self-advisory support vector machine, input the reconstructed signal into the self-advisory support vector machine for training, obtain error classification data of the reconstructed signal and assign a weight, and remove the error classification data of the reconstructed signal according to the weight to obtain a final signal; an analysis module is configured to input the final signal into the self-advisory support vector machine for prediction to obtain a contact resistance of the cable intermediate joint to be analyzed; wherein the screening module is specifically configured to: construct a self-advisory support vector machine, input the reconstructed signal into the self-advisory support vector machine for training; when If so, the weight of the misclassified data is obtained according to the preset weight calculation formula, and if the calculated weight is zero, then a decision is made. This is non-misclassified data; When , the weight of the misclassified data is obtained by a preset weight calculation formula, and the misclassified data of the reconstructed signal is removed according to the weight to obtain a final signal, wherein: ； ； In the formula, For misclassified datasets, , and It is a constant; , , , j=1, ..., N, are not belonging to Centralized training data; The preset weight calculation formula is: ； In the formula, is a weight, is a single raw data.

5. The contact resistance analysis system of a cable intermediate joint according to claim 4, characterized in that, The reconstruction module is specifically configured to: discretize the original time domain signal into a low-frequency signal and a high-frequency signal through a discrete wavelet transform; input the low-frequency signal and the high-frequency signal into a preset reconstruction formula to obtain the reconstructed signal; The preset reconstruction formula is: ； wherein is a constant, is a low frequency signal, is a high frequency signal, denotes the reconstructed signal of the inverse, denotes a low pass filter, denotes a high pass filter.

6. The contact resistance analysis system of a cable intermediate joint according to claim 4, characterized by Further comprising: a generation module; The generation module is configured to calculate a first contact resistance of a cable intermediate joint to be analyzed through a conventional calculation method, measure a second contact resistance of the cable intermediate joint through a micro-ohmmeter, and generate an analysis reference table according to the first contact resistance, the second contact resistance, and the contact resistance of the cable intermediate joint to be analyzed.

7. A contact resistance analysis device for a cable intermediate joint, characterized by, The device comprises a processor and a memory: The memory is configured to store program code and transmit the program code to the processor; The processor is configured to execute the contact resistance analysis method of the cable intermediate joint according to instructions in the program code.

8. A computer-readable storage medium, characterized in that, The computer readable storage medium is configured to store program code for executing the contact resistance analysis method of the cable intermediate joint.

Images