Cable joint health state assessment method and system based on multi-dimensional data fusion

Abstract

The invention discloses a cable joint health state assessment method and system based on multi-dimensional data fusion, and relates to the technical field of cable state assessment, and the method comprises the steps: collecting cable joint operation parameters, unifying a time sequence, and forming a two-dimensional matrix; performing spectral analysis on the operation parameters, and calculating energy characteristics of each sub-band by applying multi-stage wavelet decomposition; and constructing a Bayesian network structure, and calculating probability distribution according to the energy characteristics of the sub-bands. According to the method, spectral analysis is carried out through Fourier transform, the effect of identifying main frequency components and noise characteristics in signals is achieved, the effect of extracting energy characteristics of the signals in different frequency ranges is achieved by decomposing a two-dimensional matrix into a plurality of sub-bands by applying multi-stage wavelet decomposition, and the method has the advantages of being simple in structure and convenient to operate. By constructing a Bayesian network structure and defining feature nodes and state nodes, the health state evaluation of the cable can be dynamically updated according to data collected in real time.

Description

Technical Field

[0001] The present invention relates to the technical field of cable status assessment, and in particular to a cable joint health status assessment method and system based on multi-dimensional data fusion. Background Art

[0002] Cable joints are usually key components used in power transmission networks. Their health status is directly related to the stable operation of the system. Traditional cable joint health monitoring methods mainly rely on regular manual inspections and simple electrical parameter measurements, such as current, voltage, and temperature. However, these methods have obvious limitations: on the one hand, manual inspections are often subject to time and space constraints and cannot achieve real-time monitoring of cable joints; on the other hand, a single electrical parameter cannot fully reflect the complex health status of cable joints;

[0003] With the development of sensing technology, the fusion analysis of multi-sensor data has gradually become a research hotspot, aiming to more comprehensively reflect the operating status of cable joints through multi-dimensional data fusion. However, there are still many shortcomings. First, most traditional monitoring methods lack the comprehensive utilization of multi-dimensional data and cannot capture the complex characteristic changes of cable joints. In the face of complex environmental changes and equipment aging, it is difficult to accurately judge the true health status of cable joints. Secondly, it is impossible to efficiently extract effective information based on massive monitoring data, resulting in insufficient accuracy and timeliness of fault warning. Summary of the invention

[0004] In view of the problems existing in the above-mentioned existing cable joint health status assessment method and system based on multi-dimensional data fusion, the present invention is proposed.

[0005] Therefore, the problem to be solved by the present invention is that most traditional monitoring methods lack the comprehensive utilization of multi-dimensional data and are unable to capture the complex characteristic changes of cable joints. In the face of complex environmental changes and equipment aging, it is difficult to accurately judge the true health status of cable joints. Secondly, based on massive monitoring data, it is impossible to efficiently extract effective information, resulting in insufficient accuracy and timeliness of fault warnings.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: a cable joint health status assessment method based on multi-dimensional data fusion, which includes:

[0007] Collect the operating parameters of the cable joints, unify the time series and form a two-dimensional matrix;

[0008] Perform spectral analysis on the operating parameters, apply multi-level wavelet decomposition, and calculate the energy characteristics of each sub-band;

[0009] Construct a Bayesian network structure, calculate the probability distribution according to the energy characteristics of the sub-band, build an ANFIS adaptive network fuzzy inference model, and predict the health status of the cable based on the energy characteristics;

[0010] Define warning thresholds and set warning alarms, securely transmit and store data over wireless networks, and generate summary reports.

[0011] As a preferred solution of the cable joint health status assessment method based on multi-dimensional data fusion of the present invention, wherein: the cable joint operation parameters are collected, the time series is unified and formed into a two-dimensional matrix, including:

[0012] Install and deploy sensors along the cable joint line to collect the operating parameters of the cable joint;

[0013] Perform data cleaning and normalization based on the collected operating parameters, and unify the time series for time synchronization recording;

[0014] The operating parameters based on different sensors are organized into a two-dimensional matrix, in which each row represents a sequence of operating parameter items of different sensors, and each column represents all sensor readings at different time points.

[0015] As a preferred solution of the cable joint health status assessment method based on multi-dimensional data fusion described in the present invention, wherein: the spectrum analysis of the operating parameters is performed, multi-level wavelet decomposition is applied, and the energy characteristics of each sub-band are calculated, including:

[0016] The spectrum analysis of the operating parameters is performed using Fourier transform, which is expressed as:

[0017] ;

[0018] in represents the Fourier transform result of frequency f, represents the operating parameter value changing with time t, j represents the imaginary unit, represents the frequency angular velocity;

[0019] By converting the time domain operating parameters into frequency domain signals, the spectrum of the signal is obtained, the energy distribution of the signal at different frequencies is observed, and the highest effective frequency of the signal is determined. ;

[0020] Calculate the signal-to-noise ratio to determine the signal quality, expressed as:

[0021] ;

[0022] Where SNR stands for signal-to-noise ratio, represents the power of the signal, represents the power of the noise, represents the operation of converting power ratio into decibels;

[0023] The signal quality is determined based on the calculation of SNR, and the calculation of the decomposition level is performed, which is expressed as:

[0024] ;

[0025] Where J represents the level of wavelet decomposition, represents the sampling frequency of the signal, Indicates the highest effective frequency of the operating parameter signal;

[0026] Based on the two-dimensional matrix of the operating parameters, a multi-level wavelet decomposition is applied using filters to decompose the two-dimensional matrix into different sub-bands, which are expressed as:

[0027] ;

[0028] in Indicates the position of the j+1th layer The wavelet coefficient subbands are Represents the coefficients of the filter, which is used to extract components of specific frequencies from the signal of the previous layer. It means to sum all the values ​​of k and l, where k and l represent the index of the filter coefficients in the wavelet transform. Indicates location The wavelet coefficients of

[0029] Calculate the energy extracted from each sub-band as a feature, expressed as:

[0030] ;

[0031] in represents the sum of the energies of all wavelet coefficients in the subband, as the energy feature, The modulus of the wavelet coefficients representing the position (x, y);

[0032] Calculated based on historical data The historical mean and standard deviation of , if calculated Greater than or equal to The total energy of the monitoring is judged to be abnormal. If the calculated Less than It is determined that the total monitored energy is normal.

[0033] As a preferred solution of the cable joint health status assessment method based on multi-dimensional data fusion described in the present invention, wherein: the Bayesian network structure is constructed to calculate the probability distribution according to the energy characteristics of the sub-band, including:

[0034] Continuously calculate based on collected data And define the structure of the Bayesian network, including nodes and edges in the network, where nodes include feature nodes and state nodes, which are the energy of each subband obtained based on wavelet transform. and the health status of the corresponding cable, where the edge represents a directed edge from the feature node to the status node;

[0035] Create a conditional probability table (CPT) for each node, which describes the probability of each node state when the node state is given based on historical data or expert knowledge;

[0036] By applying the Bayesian rule, the probability distribution in the network is updated according to the latest feature data, expressed as:

[0037] ;

[0038] in represents the posterior probability of the health state given the energy characteristics, represents the probability of observing an energy signature in a specific health state, represents the prior probability of the cable health status, represents the observed energy feature probability, H represents the cable status, and F represents the energy features of different sub-bands;

[0039] Based on the historical data of the health state of slight injury and the health state of severe injury, the mean value and standard deviation of each energy feature are calculated, and the sum of the mean value and two times the standard deviation of the health state of slight injury is used as the slight injury threshold, and the sum of the mean value and three times the standard deviation of the health state of slight injury is used as the high injury threshold;

[0040] According to the updated probability distribution, the calculated probabilities of different cable health states are obtained. If the calculated probability is greater than or equal to the high-loss threshold, it is judged as severe damage. If the calculated probability is greater than or equal to the light-loss threshold and less than the high-loss threshold, it is judged as light damage. If the calculated probability is less than the light-damage threshold, it is judged as normal damage.

[0041] As a preferred solution of the cable joint health status assessment method based on multi-dimensional data fusion described in the present invention, wherein: the construction of the ANFIS adaptive network fuzzy inference model to predict the health status of the cable based on energy characteristics includes:

[0042] Construct ANFIS adaptive network fuzzy inference model, including input layer, fuzzy layer, rule layer, normalization layer and output layer;

[0043] The input layer receives energy feature data;

[0044] The fuzzy layer performs fuzzification through several membership functions, which can be expressed as:

[0045] ;

[0046] in Represents the fuzzy set of input x feature pairs The membership degree of represents the membership function center, represents the standard deviation of the membership function;

[0047] The rule layer implements the application of fuzzy rules, and the normalization layer normalizes the output of each rule. The output layer calculates the weighted output sum of each rule, and the output result is the predicted value of the cable health status.

[0048] The model is trained through the training set, and the model loss is calculated through the mean square error function, which is expressed as:

[0049] ;

[0050] Where m represents the number of samples, represents the actual health status of the i-th sample, represents the predicted value of the i-th sample, and MSE represents the calculation loss;

[0051] The model parameters are iteratively optimized using the Adam optimizer and the gradient descent method. When the model loss no longer decreases significantly during the continuous iteration process, the iteration is stopped and the model parameters are output to update the model.

[0052] After training is completed, the new energy features are input into the model to calculate the predicted value of the cable health status.

[0053] As a preferred solution of the cable joint health status assessment method based on multi-dimensional data fusion of the present invention, wherein: the definition of the early warning threshold and the setting of the early warning alarm include:

[0054] Based on Python's Pandas library, the predicted data output by the ANFIS adaptive network fuzzy inference model is cleaned and sorted, including removing outliers and filling missing data. The data visualization tool Matplotlib is used to visualize the sorted data, generate status distribution diagrams and trend diagrams, and use the sum of the mean and standard deviation of historical data as the warning threshold. If the predicted value of the cable's health status is greater than or equal to the warning threshold, it is judged that the health risk is high, and a warning alarm is triggered to notify relevant personnel.

[0055] As a preferred solution of the cable joint health status assessment method based on multi-dimensional data fusion described in the present invention, wherein: the data is securely transmitted and stored based on a wireless network to generate a summary report, including:

[0056] The prediction data output from the ANFIS model is encrypted and transmitted through the encrypted TLS / SSL protocol to ensure that the data is not stolen or tampered with during transmission. A distributed database is used to store, record and verify the integrity of the data. The Python library Matplotlib and Seaborn are used to generate reports, and the sensor monitoring data, energy characteristic data, probability distribution data, and cable health status prediction value data are entered into the report for summary. The SMTP library sends the report as an attachment to the specified email address.

[0057] Another object of the present invention is to provide a system for evaluating the health status of a cable joint based on multi-dimensional data fusion, which comprises:

[0058] Data acquisition module, collecting operating parameters of cable connectors;

[0059] The signal processing module analyzes and extracts features from operating parameters, uses Fourier transform to perform spectrum analysis on the acquired signal, applies multi-level wavelet decomposition to decompose the signal into different sub-bands, and calculates the energy features of each sub-band;

[0060] Bayesian network module, which calculates the probability distribution of cable health status based on energy characteristics;

[0061] ANFIS prediction module predicts the health status of the cable based on energy characteristics;

[0062] The early warning and alarm module defines the early warning threshold based on the mean and standard deviation of historical data and defines the triggering alarm strategy;

[0063] The data storage and notification module ensures the secure transmission and storage of data, generates and sends data analysis reports for notification.

[0064] A computer device comprises: a memory and a processor; the memory stores a computer program, and the processor implements the steps of the above-mentioned cable joint health status assessment method based on multi-dimensional data fusion when executing the computer program.

[0065] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the cable joint health status assessment method based on multi-dimensional data fusion.

[0066] The beneficial effects of the present invention are as follows: by performing spectrum analysis through Fourier transform, the main frequency components and noise characteristics in the signal can be identified; by applying multi-level wavelet decomposition, the two-dimensional matrix is ​​decomposed into multiple sub-bands, so as to extract the energy characteristics of the signal in different frequency ranges; by constructing a Bayesian network structure and defining feature nodes and status nodes, the health status assessment of the cable can be dynamically updated according to the real-time collected data; by receiving a variety of energy feature data through the ANFIS model, a comprehensive analysis of the health status of the cable can be achieved, potential fault risks can be discovered early, and the risk of unexpected downtime and accidents caused by cable faults can be reduced, thereby improving the reliability and safety of the entire cable system. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0068] Figure 1 The figure is a flow chart of the cable joint health status assessment method based on multi-dimensional data fusion.

[0069] Figure 2 It is a structural diagram of the cable joint health status assessment system based on multi-dimensional data fusion. DETAILED DESCRIPTION

[0070] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below in conjunction with the accompanying drawings.

[0071] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0072] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor is it a separate or selective embodiment that is mutually exclusive with other embodiments.

[0073] Example 1, reference Figure 1, which is the first embodiment of the present invention, and provides a cable joint health status assessment method based on multi-dimensional data fusion. The cable joint health status assessment method based on multi-dimensional data fusion includes:

[0074] S1, collects the operating parameters of the cable joints, unifies the time series and forms a two-dimensional matrix;

[0075] Preferably, the cable joint operation parameters are collected, the time series are unified and formed into a two-dimensional matrix, including:

[0076] Install and deploy sensors along the cable joint line to collect the operating parameters of the cable joint, including current, temperature, vibration, etc.

[0077] Perform data cleaning and normalization based on the collected operating parameters, and unify the time series for time synchronization recording;

[0078] The operating parameters based on different sensors are organized into a two-dimensional matrix, in which each row represents a sequence of operating parameter items of different sensors, and each column represents all sensor readings at different time points.

[0079] Through data cleaning, outliers and noise in sensor data are removed, making the data more accurate and reliable. Normalization processing eliminates the impact of different sensor data due to different dimensions, unifies the scale of the data, and unifies the time series and time synchronization records to ensure the time consistency of each sensor data. By forming the operating parameters of different sensors into a two-dimensional matrix, a comprehensive analysis of multi-dimensional data can be achieved. Each row represents the data sequence of different sensors, providing continuous monitoring of specific physical quantities; each column represents the readings of all sensors at a specific time point, providing a global view of the status of the entire system. This data structure helps to deeply analyze the operating status of the cable connector and its changing trends at different time points.

[0080] S2, perform spectrum analysis on the operating parameters, apply multi-level wavelet decomposition, and calculate the energy characteristics of each sub-band;

[0081] Preferably, the operating parameters are subjected to spectral analysis, multi-level wavelet decomposition is applied, and the energy characteristics of each sub-band are calculated, including:

[0082] The spectrum analysis of the operating parameters is performed using Fourier transform, which is expressed as:

[0083] ;

[0084] in The Fourier transform result of frequency f represents the amplitude of the signal at that frequency. represents the operating parameter value changing with time t, represents a complex exponential function, which is used to convert time domain signals to frequency domain. j represents an imaginary unit. represents the frequency angular velocity,

[0085] The purpose of spectrum analysis is to identify the main frequency components and noise characteristics in the signal. Through Fourier transform, the time domain operating parameters can be converted into frequency domain signals, thereby obtaining the signal spectrum, which shows the amplitude distribution of the collected data at each frequency and can help identify high-energy and low-energy frequency areas;

[0086] By converting the time domain operating parameters into frequency domain signals, the spectrum of the signal is obtained, the energy distribution of the signal at different frequencies is observed, and the highest effective frequency of the signal is determined. ;

[0087] Calculate the signal-to-noise ratio to determine the signal quality, expressed as:

[0088] ;

[0089] Where SNR stands for signal-to-noise ratio, represents the power of the signal, represents the power of the noise, represents the operation of converting power ratio into decibels;

[0090] The signal quality is determined based on the calculation of SNR. Generally, SNR above 20 is considered to be a good signal quality for determining the signal quality, and the calculation of the decomposition level is performed, which is expressed as:

[0091] ;

[0092] Where J represents the level of wavelet decomposition, represents the sampling frequency of the signal, Indicates the highest effective frequency of the operating parameter signal;

[0093] The derivation of this formula is based on the Nyquist sampling theorem of the operating parameter signal, which ensures that the lowest frequency component after decomposition can effectively capture the important features in the signal;

[0094] Based on the two-dimensional matrix of the operating parameters, a multi-level wavelet decomposition is applied using filters to decompose the two-dimensional matrix into different sub-bands, which are expressed as:

[0095] ;

[0096] in Indicates the position of the j+1th layer The wavelet coefficient subbands are Represents the coefficients of the filter, which is used to extract components of specific frequencies from the signal of the previous layer. It means to sum all the values ​​of k and l, where k and l represent the index of the filter coefficients in the wavelet transform. Indicates location The wavelet coefficients of

[0097] In the formula, by weighted summing of coefficients at different positions, the parameters 2k and 2l are used to ensure that the wavelet transform has appropriate scaling at each layer, so as to capture the spatial local characteristics of the signal;

[0098] Calculate the energy extracted from each sub-band as a feature, expressed as:

[0099] ;

[0100] in represents the sum of the energies of all wavelet coefficients in the subband, as the energy feature, The modulus of the wavelet coefficient representing the position (x, y). The square of the modulus is used to eliminate the phase information and retain only the energy information.

[0101] The process of extracting the sum of energy in each sub-band as a feature can provide a quantifiable measure of how much energy the signal has concentrated in a specific sub-band (frequency range). When performing cable health monitoring, certain damage or abnormal conditions may cause the signal energy at a specific frequency to increase or decrease significantly. By comparing the energy distribution at different time points or under different conditions, the health of the cable can be effectively monitored.

[0102] Calculated based on historical data The historical mean and standard deviation of , if calculated Greater than or equal to The total energy of the monitoring is judged to be abnormal. If the calculated Less than Then it is determined that the total energy monitored is normal;

[0103] The vibration modes of cables in normal operation and damaged states are different. Damage usually causes significant changes in energy in certain frequency bands. For example, looseness of a connection may lead to increased low-frequency vibration energy. By monitoring the energy changes in these specific frequency bands, potential damage locations and types can be identified. In addition to vibration, changes in temperature, pressure or load may also cause changes in signal characteristics, which will also be reflected in the energy of the wavelet coefficients. For example, increased temperature may change the physical properties of cable materials, which will appear as changes in high-frequency energy in wavelet analysis. By setting thresholds, an alarm can be issued immediately to prompt the maintenance team to conduct inspections and repairs, thereby preventing the fault from developing into a more serious problem.

[0104] Through Fourier transform spectrum analysis, the main frequency components and noise characteristics in the signal can be identified, which enables us to determine the energy distribution of cable operating parameters at different frequencies, and then identify high-energy and low-energy frequency areas. By calculating the signal-to-noise ratio (SNR), the signal quality is determined, and the effect of optimizing the selection of decomposition levels is achieved. A high SNR indicates that the signal quality is good and the noise interference is small, so fewer decomposition levels can be selected, while a low SNR indicates that more decomposition levels are needed to separate noise and useful signals to ensure that effective signal information is extracted. The decomposition level determined based on the Nyquist sampling theorem and the signal-to-noise ratio calculation ensures that the lowest frequency component after decomposition can capture important features in the signal. By applying multi-level wavelet decomposition, the two-dimensional matrix is ​​decomposed into multiple sub-bands, which achieves the effect of extracting the energy characteristics of the signal in different frequency ranges. By calculating the energy characteristics of each sub-band, the sum of the energy of the signal in a specific frequency range is obtained, which achieves the effect of quantitatively evaluating the health status of the cable. By setting energy thresholds and real-time monitoring, the effect of quickly detecting abnormal conditions and issuing timely warnings is achieved.

[0105] S3, construct a Bayesian network structure, calculate the probability distribution according to the energy characteristics of the sub-band, build an ANFIS adaptive network fuzzy inference model, and predict the health status of the cable based on the energy characteristics;

[0106] Preferably, a Bayesian network structure is constructed to calculate the probability distribution according to the energy characteristics of the sub-bands, including:

[0107] Continuously calculate based on collected data And define the structure of the Bayesian network, including nodes and edges in the network, where nodes include feature nodes and state nodes, which are the energy of each subband obtained based on wavelet transform. and the health status of the corresponding cable, which is predefined based on the performance and safety standards of the overall cable system, where the edge represents a directed edge from a feature node to a status node, which is used to represent the conditional dependency of the feature on the health status;

[0108] Create a conditional probability table (CPT) for each node, which describes the probability of each node state when the node state is given based on historical data or expert knowledge;

[0109] For example, when the decomposition level of the energy characteristic is 3, the cable status is defined as normal, slightly damaged, and severely damaged;

[0110] For normal cable status: ;

[0111] For cables with minor damage: ;

[0112] For severe cable damage: ;

[0113] in Represents the low-frequency energy characteristics of the nth sensor data, represents the intermediate frequency energy characteristics of the nth sensor data, represents the high-frequency energy characteristics of the nth sensor data, H represents the health status, and P represents the probability value;

[0114] By applying the Bayesian rule, the probability distribution in the network is updated according to the latest feature data, expressed as:

[0115] ;

[0116] in represents the posterior probability of the health state given the energy characteristics, represents the probability of observing an energy signature in a specific health state, The prior probability of the cable health status can be determined based on historical data. represents the observed energy feature probability, H represents the cable status, and F represents the energy features of different sub-bands;

[0117] Based on the historical data of the health state of slight injury and the health state of severe injury, the mean value and standard deviation of each energy feature are calculated, and the sum of the mean value and two times the standard deviation of the health state of slight injury is used as the slight injury threshold, and the sum of the mean value and three times the standard deviation of the health state of slight injury is used as the high injury threshold;

[0118] According to the updated probability distribution, the calculated probabilities of different cable health states are obtained. If the calculated probability is greater than or equal to the high-loss threshold, it is judged as severe damage. If the calculated probability is greater than or equal to the light-loss threshold and less than the high-loss threshold, it is judged as light damage. If the calculated probability is less than the light-damage threshold, it is judged as normal damage.

[0119] By constructing a Bayesian network structure and defining feature nodes and status nodes, the health status assessment of the cable can be dynamically updated based on the real-time collected data. This structure can effectively capture various characteristics, such as the impact of energy characteristics in different frequency bands on the health status, so that the system can update the status prediction in real time based on the latest data and provide a more accurate health status assessment. By applying the Bayesian rule to update the probability distribution and adjusting the posterior probability of the health status based on the latest feature data, the system can use historical data and expert knowledge to accurately infer the current cable status. Through detailed health status classification and probability judgment, the system provides strong decision support for cable maintenance.

[0120] Furthermore, an ANFIS adaptive network fuzzy inference model is constructed to predict the health status of the cable based on energy characteristics, including:

[0121] Construct ANFIS adaptive network fuzzy inference model, including input layer, fuzzy layer, rule layer, normalization layer and output layer;

[0122] The input layer receives energy feature data;

[0123] The fuzzy layer performs fuzzification through several membership functions, which can be expressed as:

[0124] ;

[0125] in Represents the fuzzy set of input x feature pairs The membership degree of represents the membership function center, represents the standard deviation of the membership function;

[0126] The rule layer implements the application of fuzzy rules, and the normalization layer normalizes the output of each rule. The output layer calculates the weighted output sum of each rule, and the output result is the predicted value of the cable health status.

[0127] The model is trained through the training set, and the model loss is calculated through the mean square error function, which is expressed as:

[0128] ;

[0129] Where m represents the number of samples, represents the actual health status of the i-th sample, represents the predicted value of the i-th sample, and MSE represents the calculation loss;

[0130] The model parameters are iteratively optimized using the Adam optimizer and the gradient descent method. When the model loss no longer decreases significantly during the continuous iteration process, the iteration is stopped and the model parameters are output to update the model.

[0131] After training is completed, the new energy features are input into the model to calculate the predicted value of the cable health status.

[0132] ANFIS model receives a variety of energy characteristic data, such as energy characteristics of different frequency sub-bands, to achieve a comprehensive analysis of the cable health status. Fuzzy processing can effectively process continuous values ​​and uncertain data, making the model more robust. The mean square error (MSE) function is used to evaluate the model's loss, and the Adam optimizer and gradient descent method are used to update the parameters to ensure efficient convergence of the model during training. The automated parameter optimization process reduces manual intervention, allowing the model to adaptively adjust according to the data and improve prediction accuracy. Through accurate health status prediction, the system can detect potential fault risks early and reduce the risk of unexpected downtime and accidents caused by cable faults, thereby improving the reliability and safety of the entire cable system.

[0133] S4, define warning thresholds and set warning alarms, securely transmit and store data based on wireless networks, and generate summary reports;

[0134] Preferably, defining warning thresholds and setting warning alarms include,

[0135] Based on Python's Pandas library, the predicted data output by the ANFIS adaptive network fuzzy inference model is cleaned and sorted, including removing outliers and filling missing data. The data visualization tool Matplotlib is used to visualize the sorted data, generate status distribution diagrams and trend diagrams, and use the sum of the mean and standard deviation of historical data as the warning threshold. If the predicted value of the cable's health status is greater than or equal to the warning threshold, it is judged that the health risk is high, and a warning alarm is triggered to notify relevant personnel.

[0136] The Pandas library is used to clean and organize data, effectively remove outliers and fill missing data to ensure data integrity and accuracy. The processed data is more reliable and provides a solid foundation for subsequent analysis and decision-making. Matplotlib is used for data visualization. The generated status distribution diagram and trend diagram can intuitively display the changes in the health status of the cable. The warning thresholds set based on the mean and standard deviation of historical data effectively build a real-time monitoring and early warning system, which can respond to potential risks in a timely manner, prevent problems from deteriorating, and ensure the safety and stability of the system.

[0137] Furthermore, based on the secure transmission and storage of data over the wireless network, a summary report is generated, including:

[0138] The prediction data output from the ANFIS model is encrypted and transmitted through the encrypted TLS / SSL protocol to ensure that the data is not stolen or tampered with during transmission. A distributed database is used to store, record and verify the integrity of the data. The Python library Matplotlib and Seaborn are used to generate reports, and the sensor monitoring data, energy characteristic data, probability distribution data, and cable health status prediction value data are entered into the report for summary. The SMTP library sends the report as an attachment to the specified email address.

[0139] The prediction data is encrypted and transmitted through the TLS / SSL protocol, which effectively prevents the data from being stolen or tampered with during transmission. The encryption mechanism ensures the confidentiality and integrity of the data, protects sensitive information, and prevents malicious attacks and information leakage. The reports generated using the Python library Matplotlib and Seaborn can intuitively display sensor monitoring data, energy characteristic data, probability distribution data, and cable health status prediction value data. The charts and analysis results contained in the report can help users clearly understand the operating status and potential risks of the cable. The generated report is sent as an attachment to the specified email address through the SMTP library, which realizes the timely delivery of information. The maintenance team and related personnel can obtain the latest monitoring and prediction results as soon as possible, so as to facilitate rapid response measures.

[0140] Example 2, reference Figure 2 , which is the second embodiment of the present invention, and which is different from the previous embodiment, provides a system for evaluating the health status of a cable joint based on multi-dimensional data fusion, including:

[0141] Data acquisition module, collecting operating parameters of cable connectors;

[0142] The signal processing module analyzes and extracts features from operating parameters, uses Fourier transform to perform spectrum analysis on the acquired signal, applies multi-level wavelet decomposition to decompose the signal into different sub-bands, and calculates the energy features of each sub-band;

[0143] Bayesian network module, which calculates the probability distribution of cable health status based on energy characteristics;

[0144] ANFIS prediction module predicts the health status of the cable based on energy characteristics;

[0145] The early warning and alarm module defines the early warning threshold based on the mean and standard deviation of historical data and defines the triggering alarm strategy;

[0146] The data storage and notification module ensures the secure transmission and storage of data, generates and sends data analysis reports for notification.

[0147] If the functions are implemented in the form of software functional units and sold or used as independent products, they 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 the 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, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program codes.

[0148] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0149] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0150] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit with a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit with a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0151] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A cable joint health status assessment method based on multi-dimensional data fusion, characterized by: include, Collect the operating parameters of the cable joints, unify the time series and form a two-dimensional matrix; Perform spectral analysis on the operating parameters, apply multi-level wavelet decomposition, and calculate the energy characteristics of each sub-band; Construct a Bayesian network structure, calculate the probability distribution according to the energy characteristics of the sub-band, build an ANFIS adaptive network fuzzy inference model, and predict the health status of the cable based on the energy characteristics; Define warning thresholds and set warning alarms, securely transmit and store data over wireless networks, and generate summary reports.

2. The cable joint health status assessment method based on multi-dimensional data fusion according to claim 1, characterized in that: The collecting cable joint operation parameters, unifying the time series and forming a two-dimensional matrix, include: Install and deploy sensors along the cable joint line to collect the operating parameters of the cable joint; Perform data cleaning and normalization based on the collected operating parameters, and unify the time series for time synchronization recording; The operating parameters based on different sensors are organized into a two-dimensional matrix, in which each row represents a sequence of operating parameter items of different sensors, and each column represents all sensor readings at different time points.

3. The cable joint health status assessment method based on multi-dimensional data fusion according to claim 2, characterized in that: The spectrum analysis of the operating parameters is performed, multi-level wavelet decomposition is applied, and the energy characteristics of each sub-band are calculated, including: The spectrum analysis of the operating parameters is performed using Fourier transform, which is expressed as: ； in represents the Fourier transform result of frequency f, represents the operating parameter value changing with time t, j represents the imaginary unit, represents the frequency angular velocity; By converting the time domain operating parameters into frequency domain signals, the spectrum of the signal is obtained, the energy distribution of the signal at different frequencies is observed, and the highest effective frequency of the signal is determined. ; Calculate the signal-to-noise ratio to determine the signal quality, expressed as: ； Where SNR stands for signal-to-noise ratio, represents the power of the signal, represents the power of the noise, represents the operation of converting power ratio into decibels; The signal quality is determined based on the calculation of SNR, and the calculation of the decomposition level is performed, which is expressed as: ； Where J represents the level of wavelet decomposition, represents the sampling frequency of the signal, Indicates the highest effective frequency of the operating parameter signal; Based on the two-dimensional matrix of the operating parameters, a multi-level wavelet decomposition is applied using filters to decompose the two-dimensional matrix into different sub-bands, which are expressed as: ； in Indicates the position of the j+1th layer The wavelet coefficient subbands are Represents the coefficients of the filter, which is used to extract components of specific frequencies from the signal of the previous layer. It means to sum all the values ​​of k and l, where k and l represent the index of the filter coefficients in the wavelet transform. Indicates location The wavelet coefficients of Calculate the energy extracted from each sub-band as a feature, expressed as: ； in represents the sum of the energies of all wavelet coefficients in the subband, as the energy feature, The modulus of the wavelet coefficients representing the position (x, y); Calculated based on historical data The historical mean and standard deviation of , if calculated Greater than or equal to The total energy of the monitoring is judged to be abnormal. If the calculated Less than It is determined that the total monitored energy is normal.

4. The cable joint health status assessment method based on multi-dimensional data fusion according to claim 3 is characterized in that: The Bayesian network structure is constructed to calculate the probability distribution according to the energy characteristics of the sub-bands, including: Continuously calculate based on collected data And define the structure of the Bayesian network, including nodes and edges in the network, where nodes include feature nodes and state nodes, which are the energy of each subband obtained based on wavelet transform. and the health status of the corresponding cable, where the edge represents a directed edge from the feature node to the status node; Create a conditional probability table (CPT) for each node, which describes the probability of each node state when the node state is given based on historical data or expert knowledge; By applying the Bayesian rule, the probability distribution in the network is updated according to the latest feature data, expressed as: ； in represents the posterior probability of the health state given the energy characteristics, represents the probability of observing an energy signature in a specific health state, represents the prior probability of the cable health status, represents the observed energy feature probability, H represents the cable status, and F represents the energy features of different sub-bands; Based on the historical data of the health state of slight injury and the health state of severe injury, the mean value and standard deviation of each energy feature are calculated, and the sum of the mean value and two times the standard deviation of the health state of slight injury is used as the slight injury threshold, and the sum of the mean value and three times the standard deviation of the health state of slight injury is used as the high injury threshold; According to the updated probability distribution, the calculated probabilities of different cable health states are obtained. If the calculated probability is greater than or equal to the high-loss threshold, it is judged as severe damage. If the calculated probability is greater than or equal to the light-loss threshold and less than the high-loss threshold, it is judged as light damage. If the calculated probability is less than the light-damage threshold, it is judged as normal damage.

5. The cable joint health status assessment method based on multi-dimensional data fusion according to claim 4, characterized in that: The ANFIS adaptive network fuzzy inference model is constructed to predict the health status of the cable based on the energy characteristics, including: Construct ANFIS adaptive network fuzzy inference model, including input layer, fuzzy layer, rule layer, normalization layer and output layer; The input layer receives energy feature data; The fuzzy layer performs fuzzification through several membership functions, which can be expressed as: ； in Represents the fuzzy set of input x feature pairs The membership degree of represents the membership function center, represents the standard deviation of the membership function; The rule layer implements the application of fuzzy rules, and the normalization layer normalizes the output of each rule. The output layer calculates the weighted output sum of each rule, and the output result is the predicted value of the cable health status. The model is trained through the training set, and the model loss is calculated through the mean square error function, which is expressed as: ； Where m represents the number of samples, represents the actual health status of the i-th sample, represents the predicted value of the i-th sample, and MSE represents the calculation loss; The model parameters are iteratively optimized using the Adam optimizer and the gradient descent method. When the model loss no longer decreases significantly during the continuous iteration process, the iteration is stopped and the model parameters are output to update the model. After training is completed, the new energy features are input into the model to calculate the predicted value of the cable health status.

6. The cable joint health status assessment method based on multi-dimensional data fusion according to claim 5, characterized in that: Defining the warning threshold and setting the warning alarm include: Based on Python's Pandas library, the predicted data output by the ANFIS adaptive network fuzzy inference model is cleaned and sorted, including removing outliers and filling missing data. The data visualization tool Matplotlib is used to visualize the sorted data, generate status distribution diagrams and trend diagrams, and use the sum of the mean and standard deviation of historical data as the warning threshold. If the predicted value of the cable's health status is greater than or equal to the warning threshold, it is judged that the health risk is high, and a warning alarm is triggered to notify relevant personnel.

7. The cable joint health status assessment method based on multi-dimensional data fusion according to claim 6, characterized in that: The secure transmission and storage of data based on the wireless network and the generation of summary reports include: The prediction data output from the ANFIS model is encrypted and transmitted through the encrypted TLS / SSL protocol to ensure that the data is not stolen or tampered with during transmission. A distributed database is used to store, record and verify the integrity of the data. The Python library Matplotlib and Seaborn are used to generate reports, and the sensor monitoring data, energy characteristic data, probability distribution data, and cable health status prediction value data are entered into the report for summary. The SMTP library sends the report as an attachment to the specified email address.

8. A system for evaluating the health status of a cable joint based on multi-dimensional data fusion according to any one of claims 1 to 7, characterized in that: include, Data acquisition module, collecting operating parameters of cable connectors; The signal processing module analyzes and extracts features from operating parameters, uses Fourier transform to perform spectrum analysis on the acquired signal, applies multi-level wavelet decomposition to decompose the signal into different sub-bands, and calculates the energy features of each sub-band; Bayesian network module, which calculates the probability distribution of cable health status based on energy characteristics; ANFIS prediction module predicts the health status of the cable based on energy characteristics; The early warning and alarm module defines the early warning threshold based on the mean and standard deviation of historical data and defines the triggering alarm strategy; The data storage and notification module ensures the secure transmission and storage of data, generates and sends data analysis reports for notification.

9. A computer device comprising: Memory and processor; The memory stores a computer program, characterized in that: when the processor executes the computer program, the steps of the cable joint health status assessment method based on multi-dimensional data fusion according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the cable joint health status assessment method based on multi-dimensional data fusion according to any one of claims 1 to 7 are implemented.

Images