Method and device for detecting degradation degree of cable based on principal component regression
By combining principal component regression (PCR) with principal component analysis (PCA) for cable degradation detection, the problems of expensive equipment and external interference in traditional methods are solved, and efficient and stable detection of cable degradation is achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-31
- Publication Date
- 2026-03-13
AI Technical Summary
Traditional cable degradation detection methods rely on expensive equipment and are susceptible to external interference, failing to meet the requirements for efficient real-time monitoring. Traditional harmonic analysis methods suffer from multicollinearity issues in high-dimensional data, leading to insufficient model stability and detection accuracy.
Principal component regression (PCR) was used in conjunction with principal component analysis (PCA) to reduce the dimensionality of high-dimensional harmonic data. Principal components were extracted by PCA and regression analysis was performed to construct a robust cable degradation detection model.
This improves the stability and detection accuracy of the model, effectively avoids multicollinearity problems, and enables stable and accurate detection of cable degradation.
Smart Images
Figure CN121658918A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of cable deterioration detection technology, and in particular to a method and apparatus for detecting the degree of cable deterioration based on principal component regression. Background Technology
[0002] As power systems become increasingly large and complex, cables play an increasingly important role in power transmission and distribution. Cable degradation, especially insulation degradation, directly affects the operational safety of power systems. Insulation degradation can lead to serious problems such as partial discharge, short circuits, and fires. Therefore, the detection and assessment of cable degradation levels have significant practical value.
[0003] Traditional cable degradation detection methods, such as partial discharge detection, dielectric loss factor measurement, and paper insulation aging diagnosis, are widely used. However, these methods typically rely on expensive specialized equipment and complex field operations, and the test results are easily affected by external environmental interference, failing to meet the demands of modern power systems for efficient, real-time monitoring.
[0004] In recent years, cable degradation detection methods based on harmonic analysis have gradually attracted attention. Harmonic current components reflect changes in the internal electrical characteristics of cables, and are therefore important indicators for judging cable degradation. However, traditional harmonic analysis methods, such as Fourier transform and statistical feature-based analysis methods, struggle to effectively handle multicollinearity issues when dealing with high-dimensional harmonic data, resulting in insufficient model stability and detection accuracy. Summary of the Invention
[0005] Therefore, it is necessary to provide a method and apparatus for detecting cable degradation based on principal component regression to address the aforementioned technical problems. This method and apparatus can at least provide stable and accurate detection of cable degradation through a pre-established principal component regression model and principal component analysis method.
[0006] In a first aspect, this application provides a method for detecting cable degradation based on principal component regression, characterized in that the method includes:
[0007] Multiple harmonic data from the current of multiple circuits are acquired, and the multiple harmonic data are converted to obtain harmonic data in multiple target formats.
[0008] For each target format of harmonic data, data preprocessing is performed on the harmonic data of that target format to obtain preprocessed target harmonic data;
[0009] For each target harmonic data, principal component analysis is performed on the target harmonic data to obtain the data matrix corresponding to the target harmonic data.
[0010] For each target harmonic data, the data matrix corresponding to the target harmonic data is input into a pre-trained target principal component regression model to obtain the predicted value of cable degradation of the circuit corresponding to the target harmonic data.
[0011] For each circuit, the predicted cable degradation level is used to determine the degradation state of the cable in that circuit.
[0012] For each circuit cable, maintenance recommendations are determined based on its degradation status.
[0013] In some embodiments, the method further includes:
[0014] Based on the cable status of multiple circuits, multiple predicted values of cable degradation, and multiple calculation parameters in the target principal component regression model, a cable degradation assessment report is generated.
[0015] The cable degradation assessment report includes the degradation status of multiple circuit cables, a graph showing the influence of harmonic components on the degradation degree, the contribution rate of principal component analysis, the regression coefficient of principal component analysis, and a degradation trend graph.
[0016] In some embodiments, data conversion is performed on the plurality of harmonic data to obtain harmonic data in multiple target formats, including:
[0017] Convert multiple harmonic data in the current of each circuit into multidimensional vector data;
[0018] Each component of the vector corresponds to a specific amplitude and phase.
[0019] In some embodiments, data preprocessing is performed on the harmonic data of each target format to obtain preprocessed target harmonic data, including:
[0020] Based on the harmonic data of the target format, the maximum value of the harmonic data component of the target format, and the minimum value of the harmonic data component of the target format, determine the normalized harmonic data.
[0021] The normalized harmonic data is denoised to obtain the denoised harmonic data.
[0022] Multiple key features are extracted from the denoised harmonic data, and a high-dimensional feature vector is determined based on these key features.
[0023] Among these key features are amplitude, phase, and total harmonic distortion rate.
[0024] In some embodiments, principal component analysis is performed on each target harmonic data to obtain the data matrix corresponding to that target harmonic data through the following steps:
[0025] The high-dimensional feature vectors are centered to obtain a centered data matrix;
[0026] Determine the covariance matrix based on the sample size and the centered data matrix;
[0027] The covariance matrix is decomposed into eigenvalues to determine multiple eigenvalues and multiple eigenvectors. Each eigenvalue is used to reflect the explanatory power of the principal component represented by the corresponding eigenvector in the data.
[0028] A predetermined number of principal components are determined from the plurality of eigenvalues, and a low-dimensional data matrix is determined based on the predetermined number of principal components. The low-dimensional data matrix is the data matrix corresponding to the target harmonic data.
[0029] In some embodiments, the principal component regression model is trained through the following steps:
[0030] Obtain the original dataset, and split the dataset and train the target principal component regression model using cross-validation to determine the target regression coefficients;
[0031] Based on the target regression coefficients, determine the trained target principal component regression model;
[0032] The principal component regression model includes the following formula:
[0033] ;
[0034] in, This is a predicted value for the degree of cable degradation. For regression coefficients, For the error term, It is a low-dimensional data matrix.
[0035] Secondly, this application also provides a cable degradation degree detection device based on principal component regression, including a harmonic data conversion module, a harmonic data processing module, a data matrix analysis module, a cable degradation degree prediction value determination module, a degradation state determination module, and a maintenance suggestion determination module. The harmonic data conversion module is used to acquire multiple harmonic data in the current of multiple circuits and convert the multiple harmonic data to obtain harmonic data in multiple target formats.
[0036] The harmonic data processing module is used to preprocess the harmonic data of each target format to obtain the preprocessed target harmonic data.
[0037] The data matrix analysis module is used to perform principal component analysis on each target harmonic data to obtain the data matrix corresponding to the target harmonic data.
[0038] The cable degradation prediction module is used to input the data matrix corresponding to each target harmonic data into a pre-trained target principal component regression model to obtain the cable degradation prediction value of the circuit corresponding to the target harmonic data.
[0039] The degradation status determination module is used to determine the degradation status of the cable in each circuit based on the predicted degradation value of the cable in that circuit.
[0040] The maintenance recommendation determination module is used to determine maintenance recommendations for each circuit cable based on its degradation status.
[0041] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:
[0042] Multiple harmonic data from the current of multiple circuits are acquired, and the multiple harmonic data are converted to obtain harmonic data in multiple target formats.
[0043] For each target format of harmonic data, data preprocessing is performed on the harmonic data of that target format to obtain preprocessed target harmonic data;
[0044] For each target harmonic data, principal component analysis is performed on the target harmonic data to obtain the data matrix corresponding to the target harmonic data.
[0045] For each target harmonic data, the data matrix corresponding to the target harmonic data is input into a pre-trained target principal component regression model to obtain the predicted value of cable degradation of the circuit corresponding to the target harmonic data.
[0046] For each circuit, the predicted cable degradation level is used to determine the degradation state of the cable in that circuit.
[0047] For each circuit cable, maintenance recommendations are determined based on its degradation status.
[0048] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the following steps:
[0049] Multiple harmonic data from the current of multiple circuits are acquired, and the multiple harmonic data are converted to obtain harmonic data in multiple target formats.
[0050] For each target format of harmonic data, data preprocessing is performed on the harmonic data of that target format to obtain preprocessed target harmonic data;
[0051] For each target harmonic data, principal component analysis is performed on the target harmonic data to obtain the data matrix corresponding to the target harmonic data.
[0052] For each target harmonic data, the data matrix corresponding to the target harmonic data is input into a pre-trained target principal component regression model to obtain the predicted value of cable degradation of the circuit corresponding to the target harmonic data.
[0053] For each circuit, the predicted cable degradation level is used to determine the degradation state of the cable in that circuit.
[0054] For each circuit cable, maintenance recommendations are determined based on its degradation status.
[0055] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, performs the following steps:
[0056] Multiple harmonic data from the current of multiple circuits are acquired, and the multiple harmonic data are converted to obtain harmonic data in multiple target formats.
[0057] For each target format of harmonic data, data preprocessing is performed on the harmonic data of that target format to obtain preprocessed target harmonic data;
[0058] For each target harmonic data, principal component analysis is performed on the target harmonic data to obtain the data matrix corresponding to the target harmonic data.
[0059] For each target harmonic data, the data matrix corresponding to the target harmonic data is input into a pre-trained target principal component regression model to obtain the predicted value of cable degradation of the circuit corresponding to the target harmonic data.
[0060] For each circuit, the predicted cable degradation level is used to determine the degradation state of the cable in that circuit.
[0061] For each circuit cable, maintenance recommendations are determined based on its degradation status.
[0062] The aforementioned method and apparatus for detecting cable degradation based on principal component regression can at least achieve stable and accurate detection of cable degradation through a pre-established principal component regression model and principal component analysis method. Attached Figure Description
[0063] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments of this application will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.
[0064] Figure 1 This is a flowchart illustrating a cable degradation detection method based on principal component regression in one embodiment.
[0065] Figure 2 This is a schematic diagram illustrating the variation of some harmonic content in one embodiment;
[0066] Figure 3 This is a schematic diagram illustrating the cumulative explained variance of the PCA components in one embodiment;
[0067] Figure 4 This is a schematic diagram of the regression coefficients of each principal component in one embodiment;
[0068] Figure 5 This is a schematic diagram comparing the actual value with the predicted value in one embodiment;
[0069] Figure 6 This is a structural block diagram of a cable degradation detection device based on principal component regression in one embodiment;
[0070] Figure 7 This is an internal structural diagram of a computer device in one embodiment.
[0071] Figure labels and descriptions:
[0072] 601. Harmonic data conversion module; 602. Harmonic data processing module; 603. Data matrix analysis module; 604. Cable degradation degree prediction module; 605. Degradation status module; 606. Maintenance suggestion module. Detailed Implementation
[0073] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.
[0074] The cable degradation detection method and apparatus based on principal component regression provided in this application can be applied to various personal computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. IoT devices can include smart speakers, smart TVs, smart air conditioners, smart in-vehicle devices, projection devices, etc. Portable wearable devices can include smartwatches, smart bracelets, head-mounted devices, etc. Head-mounted devices can include virtual reality (VR) devices, augmented reality (AR) devices, smart glasses, etc.
[0075] It should be noted that, in related technologies, the Lasso regression method is widely used in cable degradation detection. This method, by applying L1 regularization constraints to the regression coefficients, can automatically filter out key features related to the degree of degradation when processing high-dimensional harmonic data, thereby simplifying the model and enhancing its interpretability.
[0076] However, Lasso regression performs poorly in dealing with multicollinearity in high-dimensional data. When there is strong correlation between input features, model parameter estimation may be unstable, affecting prediction accuracy. Furthermore, the sparsity mechanism of Lasso regression may ignore secondary features useful for degradation assessment, potentially leading to decreased detection accuracy in complex cable environments.
[0077] Although the Lasso regression method improves detection efficiency, there is still room for improvement in handling complex data structures and enhancing model robustness.
[0078] Existing Lasso regression methods mainly suffer from the following technical problems:
[0079] (1) Multicollinearity problem: Harmonic data usually has high-dimensionality characteristics, and there is a strong correlation between different harmonic components (i.e., multicollinearity). Under multicollinearity, Lasso regression will experience unstable parameter estimation, resulting in poor prediction performance of the model. The multicollinearity problem is mainly manifested in the model being too sensitive to changes in the weights of input variables, thus affecting the robustness of the model.
[0080] (2) Insufficient model stability: Lasso regression uses L1 norm penalty for feature selection, but in practical applications, the penalty term may cause some features with important physical significance to be ignored. This neglect will cause the model to exhibit instability when facing different cable operating environments and different datasets, especially when detecting the complex and variable degree of cable degradation.
[0081] (3) Incomplete feature extraction: In high-dimensional harmonic data analysis, Lasso regression tends to select the major harmonic components with strong influence, while ignoring some minor harmonic components that may contain key information. This limitation in feature extraction may lead to insufficient accuracy in degradation detection.
[0082] To overcome the aforementioned technical problems, this invention proposes a cable degradation detection method based on the Principal Component Regression (PCR) algorithm. This method combines the advantages of Principal Component Analysis (PCA) and linear regression. PCA is used to reduce the dimensionality of high-dimensional harmonic data, extracting principal components that can explain the main variance of the data. These principal components are then used for regression analysis to construct a robust and efficient cable degradation detection model.
[0083] The core idea of the PCR method is to project high-dimensional harmonic data into a low-dimensional principal component space, which retains the most informative components of the original data. By eliminating redundant information and reducing data dimensionality, the PCR method not only improves model stability but also avoids multicollinearity problems. Furthermore, the principal components extracted by PCA often reflect the global characteristics of the harmonic data, thus ensuring the accuracy and reliability of the model.
[0084] For details, please refer to Figure 1 In one exemplary embodiment, a method for detecting cable degradation based on principal component regression is provided, including the following steps S101-S106. Wherein:
[0085] S101, acquire multiple harmonic data from the current of multiple circuits, and perform data conversion on the multiple harmonic data to obtain harmonic data in multiple target formats.
[0086] It's important to note that during operation, factors such as moisture and aging can cause insulation degradation in cables, leading to changes in their electrical characteristics. Cable degradation typically affects the harmonic components of the current, especially higher harmonics. Harmonic currents are current components generated by nonlinear loads, and their frequencies are integer multiples of the fundamental frequency. Analyzing these harmonic components can effectively assess the degree of cable degradation.
[0087] The process of converting the multiple harmonic data to obtain harmonic data in multiple target formats includes:
[0088] Convert multiple harmonic data in the current of each circuit into multidimensional vector data;
[0089] Each component of the vector corresponds to a specific amplitude and phase.
[0090] As an example, the steps to convert multiple harmonic data in the current of each circuit into multidimensional vector data include: First, during harmonic data acquisition, a high-precision harmonic analyzer can be used to collect harmonic current data of the cable under different operating conditions in real time. The data acquisition range includes the fundamental frequency (usually 50Hz) and its multiple harmonics (such as the 3rd, 5th, 7th, etc.). This harmonic data can reflect the health status of the cable. Second, the acquired harmonic data is converted into a multidimensional vector form suitable for subsequent modeling. Each component of the vector corresponds to the amplitude or phase of a specific harmonic. For example, for a circuit containing... For cables transmitting subharmonics, the collected data can be represented as follows:
[0091] ;
[0092] in, Indicates the first The amplitude or phase of the subharmonics. Finally, the collected data can be labeled, including multidimensional vector data with known degradation levels, which will be used for subsequent model training.
[0093] S102, For each target format of harmonic data, perform data preprocessing on the target format of harmonic data to obtain preprocessed target harmonic data.
[0094] Before performing regression analysis, the collected harmonic data needs to be preprocessed to improve the model's stability and prediction accuracy. Data preprocessing steps include normalization, denoising, and feature extraction.
[0095] Specifically, data preprocessing is performed on the harmonic data of each target format to obtain preprocessed target harmonic data, including:
[0096] Based on the harmonic data of the target format, the maximum value of the harmonic data component of the target format, and the minimum value of the harmonic data component of the target format, determine the normalized harmonic data.
[0097] The normalized harmonic data is denoised to obtain the denoised harmonic data.
[0098] Multiple key features are extracted from the denoised harmonic data, and a high-dimensional feature vector is determined based on these key features.
[0099] Among these key features are amplitude, phase, and total harmonic distortion rate.
[0100] Specifically, because there can be significant differences in magnitude between the amplitude and phase of harmonic components, harmonic data needs to be normalized to eliminate the influence of dimensions. The standard formula for normalization is:
[0101] ;
[0102] Where x is the original data, It is normalized data. and These are the minimum and maximum values of the component, respectively. Through normalization, the values of all harmonic components are mapped to the same scale range (e.g., [0, 1]).
[0103] Specifically, cable harmonic data may be affected by environmental noise during acquisition, thus requiring denoising processing. Commonly used denoising methods include wavelet transform and fast Fourier transform (FFT), which can effectively filter out high-frequency noise and retain the most valuable harmonic information for degradation assessment.
[0104] Specifically, key features, such as the amplitude, phase, and total harmonic distortion (THD) of each harmonic, are extracted from the normalized and denoised data. The extracted data forms a high-dimensional feature vector, which is used for principal component analysis (PCA) and principal component regression (PCR).
[0105] S103. For each target harmonic data, perform principal component analysis on the target harmonic data to obtain the data matrix corresponding to the target harmonic data.
[0106] Principal Component Analysis (PCA) is a commonly used data dimensionality reduction technique that aims to map high-dimensional data to a low-dimensional space through linear transformation, while preserving as much variance information as possible from the original data. The basic principle of PCA is to determine the principal components of the data—that is, the linear combinations that can explain the data variance—using the covariance matrix.
[0107] The following steps are used to perform principal component analysis on each target harmonic data to obtain the corresponding data matrix:
[0108] The high-dimensional feature vectors are centered to obtain a centered data matrix;
[0109] Determine the covariance matrix based on the sample size and the centered data matrix;
[0110] The covariance matrix is decomposed into eigenvalues to determine multiple eigenvalues and multiple eigenvectors. Each eigenvalue is used to reflect the explanatory power of the principal component represented by the corresponding eigenvector in the data.
[0111] A predetermined number of principal components are determined from the plurality of eigenvalues, and a low-dimensional data matrix is determined based on the predetermined number of principal components. The low-dimensional data matrix is the data matrix corresponding to the target harmonic data.
[0112] Specifically, centering the high-dimensional feature vector can be achieved by subtracting the mean of each feature, making the mean of the data zero.
[0113] ;
[0114] in, It is the mean vector of the features. It is a centralized data matrix.
[0115] Specifically, the covariance matrix can be determined using the following formula:
[0116] ;
[0117] in, It is the sample size. It is the transpose of a centralized data matrix.
[0118] Specifically, eigenvalue decomposition can be performed using the following formula:
[0119] ;
[0120] Among them, eigenvalues This reflects the interpretability of the principal components represented by the corresponding eigenvectors in the data.
[0121] Specifically, you can choose the top performers that explain the largest variance. These principal components form a new low-dimensional data space. These principal components are linear combinations of the data in the new space, capable of preserving most of the information.
[0122] ;
[0123] in, It is from the front A matrix composed of eigenvectors It is the data matrix after dimensionality reduction.
[0124] In this way, PCA can project the original high-dimensional data into a lower-dimensional space, effectively reducing data redundancy and avoiding interference from multicollinearity.
[0125] S104. For each target harmonic data, input the data matrix corresponding to the target harmonic data into the pre-trained target principal component regression model to obtain the predicted value of the cable degradation degree of the circuit corresponding to the target harmonic data.
[0126] Here, Principal Component Regression (PCR) is a regression analysis method that combines PCA with linear regression. By first performing PCA to reduce the dimensionality of the data, and then performing linear regression in the reduced principal component space, the stability and prediction accuracy of the model when dealing with high-dimensional data can be improved.
[0127] The principal component regression model can be trained using the following steps:
[0128] Obtain the original dataset, and split the dataset and train the target principal component regression model using cross-validation to determine the target regression coefficients;
[0129] Based on the target regression coefficients, determine the trained target principal component regression model.
[0130] The principal component regression model includes the following formula:
[0131] ;
[0132] in, This is a predicted value for the degree of cable degradation. For regression coefficients, For the error term, It is a low-dimensional data matrix.
[0133] As an example, to avoid overfitting, cross-validation is used to evaluate and adjust the model. Cross-validation involves repeatedly splitting the dataset, training the model, and validating it on unseen data to ultimately determine the optimal regression coefficients. This ensures that the model has good generalization ability.
[0134] Here, a trained PCR model can also be used to analyze newly acquired cable harmonic data and output predicted values for the degree of cable degradation. The accuracy and robustness of the PCR model can be evaluated by comparing the model's predictions with actual conditions.
[0135] Thus, the cable degradation detection method based on principal component regression can not only effectively reduce dimensionality through PCA, but also achieve quantitative assessment of cable degradation through linear regression analysis, ensuring the stability and accuracy of the model when processing high-dimensional harmonic data.
[0136] S105, for each circuit, based on the predicted value of cable degradation, determine the degradation state of the cable in that circuit.
[0137] Here, after the PCR model predicts the degree of cable degradation, it can generate a detailed cable degradation assessment report. The report includes quantification of degradation, visualization analysis, and maintenance recommendations.
[0138] The degree of cable degradation can be quantitatively assessed using predictions from PCR models. This assessment helps power system maintenance personnel understand the cable's health status in a timely manner and take appropriate maintenance measures.
[0139] The visualization can be used to show the impact of each harmonic component on the degree of degradation, as well as the prediction results of the PCR model. The visualization results include the contribution rate of principal components, the magnitude of regression coefficients, and degradation trend graphs, which help to intuitively understand the health status of the cable.
[0140] S106, For each circuit cable, based on the condition of the circuit cable, determine the maintenance recommendations for that circuit cable.
[0141] Here, based on the deterioration assessment results, cable maintenance recommendations can be provided, such as recommending the replacement of severely deteriorated cables and arranging regular monitoring, to ensure the safe and stable operation of the power system.
[0142] In an exemplary embodiment, to verify the effectiveness of the method of the present invention in practical applications, a field experiment was conducted on a 10kV damp and aged cable. The experiment lasted 207 days, during which harmonic data of the cable were collected daily, and the test frequencies included the 2nd to 11th harmonics. In each test, harmonic content data of the cable at different aging time points were collected and recorded. To ensure the representativeness and continuity of the experimental data, the collected data were systematically summarized and analyzed in chronological order. Some harmonic content data are shown below. Figure 2 As shown.
[0143] To further explore the correlation between harmonic data and cable degradation, principal component analysis (PCA) was performed on the preprocessed harmonic data. The main purpose of PCA is to simplify the data structure and extract characteristic components that significantly contribute to the cable degradation process through dimensionality reduction techniques. In this process, the contribution rate of each principal component to the total variance was calculated and analyzed, and finally, principal components explaining at least 95% of the total variance were retained. These principal components represent the most important information in the original harmonic data and are helpful for further regression analysis. Figure 3 This shows the cumulative explained variance of the PCA components. From Figure 3 As can be seen, the cumulative variance explained by the first 6 principal components exceeds 95%, therefore, the first 6 principal components are selected as the basis for subsequent analysis.
[0144] After determining the number of principal components, principal component regression (PCR) analysis was performed based on these six principal components. The PCR method combines principal component analysis and linear regression, which can reduce the impact of multicollinearity on the regression model and improve the model's stability and predictive ability. Through analysis of the experimental data, the regression coefficients of each principal component were obtained, as shown in the following figures. Figure 4 As shown.
[0145] Furthermore, based on these regression coefficients, a regression equation for the degree of deterioration of damp and aged cables is derived, as shown in the following expression:
[0146] ;
[0147] in, Indicates the degree of degradation. Indicates the first One principal component.
[0148] The equation shows that the influence of each principal component on the degree of degradation varies significantly, with some components contributing positively and others exhibiting negative effects. To verify the predictive power of this regression equation, a detailed comparison was made between the predicted degradation values calculated based on the equation and the actual degradation levels observed in testing. The comparison results are as follows: Figure 5 As shown.
[0149] like Figure 5 As shown in the figure, there is a high degree of fit between the actual and predicted values, indicating that the PCR model can accurately predict the degradation status of the cable.
[0150] This experiment demonstrates that Principal Component Regression (PCR) can effectively extract key features and establish accurate prediction models when processing harmonic data from damp and aging cables. This method not only reduces data dimensionality and improves model stability but also significantly enhances the accuracy of degradation prediction. This experimental verification provides a scientific basis for online cable monitoring and condition assessment, and also offers strong support for the maintenance and replacement of degraded cables in practical engineering projects.
[0151] This application combines the advantages of principal component analysis (PCA) and linear regression. PCA is used to reduce the dimensionality of high-dimensional harmonic data and extract principal components that can explain the main variance of the data. Then, these principal components are used for regression analysis to construct a robust and efficient cable degradation detection model.
[0152] The core idea of the PCR method is to project high-dimensional harmonic data into a low-dimensional principal component space, which retains the most informative components of the original data. By eliminating redundant information and reducing data dimensionality, the PCR method not only improves model stability but also avoids multicollinearity problems. Furthermore, the principal components extracted by PCA often reflect the global characteristics of the harmonic data, thus ensuring the accuracy and reliability of the model.
[0153] It should be understood that although the steps in the flowcharts of the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.
[0154] Based on the same inventive concept, this application also provides a cable degradation detection device based on principal component regression for implementing the cable degradation detection method based on principal component regression described above. The solution provided by this device is similar to the solution described in the above method. Therefore, the specific limitations of one or more embodiments of the cable degradation detection device based on principal component regression provided below can be found in the limitations of the cable degradation detection method based on principal component regression described above, and will not be repeated here.
[0155] Please refer to Figure 6 In one exemplary embodiment, a cable degradation degree detection device based on principal component regression is provided, including: a harmonic data conversion module 601, a harmonic data processing module 602, a data matrix analysis module 603, a cable degradation degree prediction value determination module 604, a degradation state determination module 605, and a maintenance suggestion determination module 606. The harmonic data conversion module is used to acquire multiple harmonic data in the current of multiple circuits and convert the multiple harmonic data to obtain harmonic data in multiple target formats.
[0156] The harmonic data processing module is used to preprocess the harmonic data of each target format to obtain the preprocessed target harmonic data.
[0157] The data matrix analysis module is used to perform principal component analysis on each target harmonic data to obtain the data matrix corresponding to the target harmonic data.
[0158] The cable degradation prediction module is used to input the data matrix corresponding to each target harmonic data into a pre-trained target principal component regression model to obtain the cable degradation prediction value of the circuit corresponding to the target harmonic data.
[0159] The degradation status determination module is used to determine the degradation status of the cable in each circuit based on the predicted degradation value of the cable in that circuit.
[0160] The maintenance recommendation determination module is used to determine maintenance recommendations for each circuit cable based on its degradation status.
[0161] Each module in the aforementioned cable degradation detection device based on principal component regression can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of a computer device in software form, so that the processor can call and execute the corresponding operations of each module.
[0162] In an exemplary embodiment, a computer device is provided, which may be a terminal. The computer device includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor of the computer device provides computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The input / output interface of the computer device is used for exchanging information between the processor and external devices. The communication interface of the computer device is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, Near Field Communication (NFC), or other technologies. When the computer program is executed by the processor, it implements a cable degradation detection method based on principal component regression. The display unit of the computer device is used to form a visually visible image and may be a display screen, a projection device, or a virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.
[0163] Those skilled in the art will understand that Figure 7 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.
[0164] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:
[0165] Multiple harmonic data from the current of multiple circuits are acquired, and the multiple harmonic data are converted to obtain harmonic data in multiple target formats.
[0166] For each target format of harmonic data, data preprocessing is performed on the harmonic data of that target format to obtain preprocessed target harmonic data;
[0167] For each target harmonic data, principal component analysis is performed on the target harmonic data to obtain the data matrix corresponding to the target harmonic data.
[0168] For each target harmonic data, the data matrix corresponding to the target harmonic data is input into a pre-trained target principal component regression model to obtain the predicted value of cable degradation of the circuit corresponding to the target harmonic data.
[0169] For each circuit, the predicted cable degradation level is used to determine the degradation state of the cable in that circuit.
[0170] For each circuit cable, maintenance recommendations are determined based on its degradation status.
[0171] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, performs the following steps:
[0172] Multiple harmonic data from the current of multiple circuits are acquired, and the multiple harmonic data are converted to obtain harmonic data in multiple target formats.
[0173] For each target format of harmonic data, data preprocessing is performed on the harmonic data of that target format to obtain preprocessed target harmonic data;
[0174] For each target harmonic data, principal component analysis is performed on the target harmonic data to obtain the data matrix corresponding to the target harmonic data.
[0175] For each target harmonic data, the data matrix corresponding to the target harmonic data is input into a pre-trained target principal component regression model to obtain the predicted value of cable degradation of the circuit corresponding to the target harmonic data.
[0176] For each circuit, the predicted cable degradation level is used to determine the degradation state of the cable in that circuit.
[0177] For each circuit cable, maintenance recommendations are determined based on its degradation status.
[0178] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.
[0179] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.
[0180] The above embodiments are merely illustrative of several implementation methods of this application, and their descriptions are relatively specific and detailed. However, they should not be construed as limiting the scope of this application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.
Claims
1. A method for detecting cable degradation based on principal component regression, characterized in that, The method includes: Multiple harmonic data from the current of multiple circuits are acquired, and the multiple harmonic data are converted to obtain harmonic data in multiple target formats. For each target format of harmonic data, data preprocessing is performed on the harmonic data of that target format to obtain preprocessed target harmonic data; For each target harmonic data, principal component analysis is performed on the target harmonic data to obtain the data matrix corresponding to the target harmonic data. For each target harmonic data, the data matrix corresponding to the target harmonic data is input into a pre-trained target principal component regression model to obtain the predicted value of cable degradation of the circuit corresponding to the target harmonic data. For each circuit, the predicted cable degradation level is used to determine the degradation state of the cable in that circuit. For each circuit cable, maintenance recommendations are determined based on its degradation status.
2. The cable degradation detection method according to claim 1, characterized in that, The method further includes: Based on the cable status of multiple circuits, multiple predicted values of cable degradation, and multiple calculation parameters in the target principal component regression model, a cable degradation assessment report is generated. The cable degradation assessment report includes the degradation status of multiple circuit cables, a graph showing the influence of harmonic components on the degradation degree, the contribution rate of principal component analysis, the regression coefficient of principal component analysis, and a degradation trend graph.
3. The cable degradation detection method according to claim 1, characterized in that, The multiple harmonic data are converted to obtain harmonic data in multiple target formats, including: Convert multiple harmonic data in the current of each circuit into multidimensional vector data; Each component of the vector corresponds to a specific amplitude and phase.
4. The cable degradation detection method according to claim 1, characterized in that, Data preprocessing is performed on the harmonic data of each target format to obtain preprocessed target harmonic data, including: Based on the harmonic data of the target format, the maximum value of the harmonic data component of the target format, and the minimum value of the harmonic data component of the target format, determine the normalized harmonic data. The normalized harmonic data is denoised to obtain the denoised harmonic data. Multiple key features are extracted from the denoised harmonic data, and a high-dimensional feature vector is determined based on these key features. Among these key features are amplitude, phase, and total harmonic distortion rate.
5. The cable degradation detection method according to claim 1, characterized in that, The following steps are used to perform principal component analysis on each target harmonic data point to obtain the corresponding data matrix: The high-dimensional feature vectors are centered to obtain a centered data matrix; Determine the covariance matrix based on the sample size and the centered data matrix; The covariance matrix is decomposed into eigenvalues to determine multiple eigenvalues and multiple eigenvectors. Each eigenvalue is used to reflect the explanatory power of the principal component represented by the corresponding eigenvector in the data. A predetermined number of principal components are determined from the plurality of eigenvalues, and a low-dimensional data matrix is determined based on the predetermined number of principal components. The low-dimensional data matrix is the data matrix corresponding to the target harmonic data.
6. The cable degradation detection method according to claim 1, characterized in that, Train the principal component regression model using the following steps: Obtain the original dataset, and split the dataset and train the target principal component regression model using cross-validation to determine the target regression coefficients; Based on the target regression coefficients, determine the trained target principal component regression model; The principal component regression model includes the following formula: ； in, This is a predicted value for the degree of cable degradation. For regression coefficients, For error terms, It is a low-dimensional data matrix.
7. A cable degradation detection device based on principal component regression, characterized in that, The device includes: The harmonic data conversion module is used to acquire multiple harmonic data from the current of multiple circuits and convert the multiple harmonic data to obtain harmonic data in multiple target formats. The harmonic data processing module is used to preprocess the harmonic data of each target format to obtain the preprocessed target harmonic data. The data matrix analysis module is used to perform principal component analysis on each target harmonic data to obtain the data matrix corresponding to the target harmonic data. The cable degradation prediction module is used to input the data matrix corresponding to each target harmonic data into a pre-trained target principal component regression model to obtain the cable degradation prediction value of the circuit corresponding to the target harmonic data. The degradation status determination module is used to determine the degradation status of the cable in each circuit based on the predicted degradation value of the cable in that circuit. The maintenance recommendation determination module is used to determine maintenance recommendations for each circuit cable based on its degradation status.
8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.
9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.
10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.