System and method for evaluating aging state of superconducting cable based on mixed data

Abstract

The invention discloses a superconducting cable aging state evaluation system and method based on mixed data, and the system comprises a data analysis module, an aging evaluation module, a physicochemical association module and a report generation module which are in communication connection. Wherein the data analysis module comprises a simulation sub-module, a data acquisition sub-module and a data processing sub-module connected with the data acquisition sub-module, and in the aging state evaluation process of the aging evaluation module, superconducting cable system model parameters are output and calibrated through a physicochemical association model; and then an aging state evaluation fitting function is output through the data processing sub-module, and an aging state evaluation report is automatically generated by the report generation module. According to the method, the operation state of the superconducting cable is evaluated by analyzing the mixed data, aging analysis of the superconducting cable system under multiple working conditions is carried out, and decision support is provided for system maintenance.

Description

Technical Field

[0001] The present invention relates to the technical field of power operation and maintenance, and in particular to a superconducting cable aging state evaluation system and method based on hybrid data. Technical Background

[0002] With the continuous growth of the demand for electric energy, especially the continuous increase in the load density of big cities, the superconducting cable power transmission technology has shown unique performance advantages in solving the tension of power transmission channels, improving transmission efficiency and reducing environmental impacts. Utilizing the characteristic of zero resistance of superconducting materials in the superconducting state, superconducting cables can achieve high-density and low-loss electric energy transmission, significantly enhancing the power transmission capacity and reducing energy losses. This technology has important application value in short-distance large-current transmission scenarios such as from generators to transformers, new energy power source output, from substations to urban power grid ports, and power plants and substations, and also shows great potential in the power transmission of large or extra-large cities.

[0003] However, as the superconducting cable technology gradually moves towards long-term engineering practical applications, the challenges it faces are becoming increasingly prominent. The superconducting cable system has a unique system composition and a harsh operating environment. These factors, including electrical parameters and non-electrical parameters, and the significantly increased importance of non-electrical parameters, make it difficult to directly apply traditional cable detection and test technologies. In addition, due to the lack of operating experience, the evaluation technology of the superconducting cable system is not yet perfect. Superconducting cables not only need to work stably at a low temperature, such as -196°C for a long time, but also need to cope with various complex working conditions such as large current fluctuations, instantaneous high loads, and the output of new energy power with randomness and periodicity. Coupled with the particularity and complex structure of the superconducting cable itself, the material properties are significantly different at low temperatures and normal temperatures, and the cooling system is a key component, and the power grid has little operating and research experience, and no aging research has been carried out. These factors cover changes in mechanical, thermal, and insulation properties, further exacerbating the difficulty of aging evaluation and characterization.

[0004] In addition, the superconducting cable system has its own particularity, and the accidents caused by the short-board effect have a huge impact. Therefore, it is necessary to integrate and synergistically consider the aging state evaluation and its characterization technology of multiple parameters of the superconducting cable system to accurately grasp the aging process of the system and each component, and conduct maintenance in a timely manner to ensure the long-term safe, stable, and efficient operation of the superconducting cable power transmission system. Summary of the Invention

[0005] In order to overcome the difficulty in evaluating the aging state of superconducting cable systems in the field of power operation and maintenance during long-term operation in the above-mentioned existing technologies, the applicant proposes a superconducting cable aging state evaluation system and method based on hybrid data, which analyzes the hybrid data to evaluate the operating state of superconducting cables, conducts aging analysis of superconducting cable systems under multiple working conditions, and provides decision-making support for system maintenance.

[0006] One of the technical solutions to achieve the above object is: a superconducting cable aging state evaluation system based on hybrid data, characterized in that: the system includes a data analysis module, an aging evaluation module, a physical-chemical correlation module, and a report generation module that are communicatively connected, where: the data analysis module includes a simulation sub-module, a data acquisition sub-module, and a data processing sub-module connected thereto. During the aging state evaluation process carried out by the aging evaluation module, the model parameters of the superconducting cable system are output and calibrated through a physical-chemical correlation model, and then an aging state evaluation fitting function is output through the data processing sub-module, and the aging state evaluation report is automatically generated by the report generation module.

[0007] In the above-mentioned superconducting cable aging state evaluation system based on hybrid data, the simulation sub-module is used to simulate the operating state parameters of the superconducting cable system under fault conditions; the data acquisition sub-module includes a number of sensors for real-time monitoring of the operating state parameters and environmental state parameters of the superconducting cable system; the data processing sub-module is used to clean, standardize, and extract features from the collected data; the physical-chemical correlation module is used to calibrate the correlation data of the aging state of the superconducting cable system; the aging evaluation module is used to calculate the aging function of the superconducting cable system and evaluate the real-time aging degree; the report generation module is used to generate the aging state evaluation report of the superconducting cable system.

[0008] In the above-mentioned superconducting cable aging state evaluation system based on hybrid data, the simulation sub-module adopts a preset fault condition model to simulate the change trends of the current, voltage, and body temperature parameters of the superconducting cable system.

[0009] In the above-mentioned superconducting cable aging state evaluation system based on hybrid data, the number of sensors in the data acquisition sub-module are current sensors, voltage sensors, temperature sensors, and vacuum sensors, which are used to real-time monitor the current value, voltage value, body temperature value, environmental temperature value, and vacuum value of the superconducting cable system.

[0010] In the above-mentioned superconducting cable aging state evaluation system based on hybrid data, the data processing sub-module removes abnormal data by using a data cleaning algorithm and unifies the data of different parameters to the same dimension by using a standardization algorithm.

[0011] In the above superconducting cable aging state evaluation system based on hybrid data, the physical-chemical correlation module uses physical-chemical analysis methods to establish the correlation between the operating state parameters and the aging state of the superconducting cable system, and calibrates the correlation data.

[0012] In the above superconducting cable aging state evaluation system based on hybrid data, the aging evaluation module uses a preset aging function, combines the real-time collected operating state parameters and environmental state parameters, calculates the real-time aging degree of the superconducting cable system, and divides the aging level.

[0013] The second technical solution to achieve the above object is: a superconducting cable aging state evaluation method based on the above system, including the following steps:

[0014] Step ①: Construct a multivariate Weibull distribution model F(X);

[0015] Step ②: Obtain the joint probability density function f(x) by taking partial derivatives;

[0016] Step ③: Organize the log-likelihood function log(f(x));

[0017] Step ④: Calculate the minimum value of the negative log-likelihood function to obtain the estimated values of the model parameters;

[0018] Step ⑤: Fit the model parameters through the multi-layer perceptron neural network algorithm;

[0019] Step ⑥: Obtain the aging function expression and evaluate the aging state of the superconducting cable system.

[0020] In the above superconducting cable aging state evaluation method, the implementation process of step ① is as follows:

[0021] K characteristic parameters X(t) = [X1(t), …, X K (t)] follow a multivariate Weibull distribution. Construct a degradation distribution function F(X) with K characteristic parameters. Then the reliability function of the superconducting cable system at time is:

[0022]

[0023] In the formula, η(t) = (η1(t), …, η K (t)) is the scale parameter; β(t) = (β1(t), …, β K (t)) is the shape parameter: θ(t) = (θ1(t), …, θ K (t)) is the correlation parameter; D f (t) = (D f1 (t), …, D fK(t) is the aging evaluation index; g(t) is the life parameter.

[0024] In the above superconducting cable aging state evaluation method, the implementation process of step ⑤ is as follows:

[0025] Construct a multi-layer perceptron neural network model, including an input layer, a hidden layer, and an output layer. The superconducting cable system model parameters are input through the input layer, and the aging state evaluation fitting function is output through the output layer. Use the backpropagation algorithm to train the neural network model and adjust the weights and bias terms.

[0026] The beneficial effects of the present invention are as follows: By integrating electrical parameters such as current and voltage and non-electrical parameters such as temperature, vacuum degree, and material properties, the present invention comprehensively evaluates the aging state of the superconducting cable system, solves the limitation of traditional methods relying on a single data source, and significantly improves the comprehensiveness and accuracy of the evaluation; Based on physical and chemical analysis and data-driven methods, a multi-parameter correlation model is constructed, which can meet the aging evaluation requirements of the superconducting cable system under complex working conditions such as extreme temperature, instantaneous overload, and new energy power transmission, providing a scientific basis for the safety and stability of the system in a multi-operating environment; By accurately calculating the aging function and dividing the aging level, it provides scientific decision-making support for the maintenance and replacement of the superconducting cable system, effectively reducing the maintenance cost, extending the service life of the system, ensuring its long-term safe, stable, and efficient operation, and promoting the wide application of superconducting cable technology. Description of the Drawings

[0027] Figure 1 It is the structural block diagram of the system of the present invention;

[0028] Figure 2 It is the flow chart of the collaborative work of each module in the system of the present invention;

[0029] Figure 3 It is the schematic diagram of the neural network algorithm for data calibration by the physical and chemical correlation module in the present invention;

[0030] Figure 4 It is one of the schematic diagrams of the comparison results between the true value and the predicted value of the aging evaluation training set in the present invention;

[0031] Figure 5 It is the second schematic diagram of the comparison results between the true value and the predicted value of the aging evaluation test set in the present invention. Detailed Embodiments

[0032] The present invention will be described in detail below with reference to the drawings and specific embodiments.

[0033] This embodiment provides a superconducting cable aging state evaluation system based on hybrid data, which can comprehensively evaluate the aging state of the superconducting cable system by integrating hybrid data, solve the limitations of a single data source, and improve the comprehensiveness of the evaluation. Based on physical and chemical analysis and data-driven methods, it can meet the aging evaluation requirements of the superconducting cable system under complex working conditions and has a positive effect on the operation and maintenance of the superconducting cable system and the power system.

[0034] As Figure 1 shown, the system includes a data analysis module, an aging evaluation module, a physical and chemical correlation module, and a report generation module that are communicatively connected. The data analysis module includes a simulation sub-module, a data acquisition sub-module, and a data processing sub-module. During the aging state evaluation process, the aging evaluation module outputs the model parameters of the superconducting cable system and calibrates them through the physical and chemical correlation model. Then, the data processing sub-module outputs the aging state evaluation fitting function, and finally, the report generation module automatically generates an aging state evaluation report. Among them:

[0035] The simulation sub-module is used to simulate the operating state parameters of the superconducting cable system under fault conditions. Based on a preset fault condition model, it can simulate the change trends of the current, voltage, and body temperature parameters of the superconducting cable system.

[0036] The data acquisition sub-module includes various sensors such as current sensors, voltage sensors, temperature sensors, and vacuum sensors, which are used to real-time monitor the current value, voltage value, body temperature value, ambient temperature value, and vacuum value of the superconducting cable system.

[0037] The data processing sub-module removes abnormal data by using a data cleaning algorithm and unifies the data of different parameters to the same dimension by using a standardization algorithm.

[0038] The physical and chemical correlation module is used to calibrate the correlation data of the aging state of the superconducting cable system. By using a physical and chemical analysis method, it establishes the correlation relationship between the operating state parameters and the aging state of the superconducting cable system and calibrates the correlation data.

[0039] The aging evaluation module is used to calculate the aging function of the superconducting cable system and evaluate the real-time aging degree. Based on a preset aging function, it combines the real-time collected operating state parameters and environmental state parameters to calculate the real-time aging degree of the superconducting cable system and divides the aging levels.

[0040] The report generation module is used to generate an aging state evaluation report of the superconducting cable system.

[0041] As Figure 2 shown, the specific implementation steps of each module of the above system are as follows:

[0042] Step ①: Construct a multivariate Weibull distribution model F(X)

[0043] K characteristic parameters \(X(t)=[X_1(t),\ldots,X\) K (t)] follow a multivariate Weibull distribution. A degradation distribution function \(F(X)\) with \(K\) characteristic parameters is constructed. Then, the reliability function of the superconducting cable system at time \(t\) is

[0044]

[0045] where \(\eta(t)=(\eta_1(t),\ldots,\eta\) K (t)) is the scale parameter; \(\beta(t)=(\beta_1(t),\ldots,\beta\) K (t)) is the shape parameter; \(\theta(t)=(\theta_1(t),\ldots,\theta\) K (t)) is the correlation parameter; \(D\) f (t)=(D f1 (t),\ldots,D_f K (t)) is the aging evaluation index; \(g(t)\) is the life parameter.

[0046] Through the simulation sub-module of the data analysis module, based on a preset fault condition model, the operating state parameters (including current, voltage, and body temperature) of the superconducting cable system under extreme conditions are simulated to generate simulation data, which is used to supplement the deficiencies of actual monitoring data, especially the data under fault conditions; through the data acquisition sub-module of the data analysis module, the operating state parameters and environmental state parameters of the superconducting cable system are monitored in real time using current sensors, voltage sensors, temperature sensors, and vacuum sensors; by cleaning the simulation data and monitoring data, outliers and noise are removed to ensure data quality, \(K\) parameters are preferably selected as \(X(t)\), and a standardization algorithm is used to unify the data of these parameters to the same dimension.

[0047] Step ②: Obtain the joint probability density function \(f(x)\) by taking partial derivatives

[0048] The degradation distribution function \(F(X)\) is a multivariate cumulative distribution function, representing the probability that all components of \(K\) characteristic parameters \(X(t)=[X_1(t),\ldots,X\) K (t)] are simultaneously less than or equal to a certain value. The joint probability density distribution function \(f(x)\) can be obtained by taking partial derivatives of the degradation distribution function \(F(X)\).

[0049] Step ③: Organize the log-likelihood function \(\log(f(x))\)

[0050] The maximum likelihood estimation method is used to estimate the parameters of the multivariate Weibull distribution to obtain the estimated values of the model parameters. Among them, the maximum likelihood estimation method is a parameter estimation method. For \(K\) characteristic parameters \(X(t)=[X_1(t),\ldots,X\) K(t)]Find a set of parameter values that maximize the probability (or probability density) of the sample data under these parameter values. For the multivariate Weibull distribution, the goal of the maximum likelihood estimation method is to estimate the unknown parameters in the distribution from the sample data. Since the calculation of the likelihood function in product form is complex, the log-likelihood function is usually taken.

[0051] Step ④: Calculate the minimum value of the negative log-likelihood function to obtain the estimated values of the model parameters

[0052] The minimum value of the negative log-likelihood function corresponds to the maximum value of the likelihood function, transforming the problem of maximizing the likelihood function into a minimization problem. Calculate the gradient of the negative log-likelihood function, and use an optimization algorithm to iteratively update the parameter values, gradually reducing the value of the negative log-likelihood function. When the change amount is less than the preset threshold or the maximum number of iterations is reached, the estimated values of the model parameters are obtained.

[0053] Step ⑤: Fit the model parameters through the multi-layer perceptron neural network algorithm

[0054] To further calibrate the model parameters, a multi-layer perceptron (MLP) model is constructed, including an input layer, a hidden layer, and an output layer. The model parameters of the superconducting cable system are input through the input layer, and the aging state evaluation fitting function is output through the output layer, as Figure 3 shown. Use the backpropagation algorithm to train the MLP model and adjust the weights and bias terms. Among them, the optimization goal is to minimize the loss function (mean square error). Use the ReLU function as the activation function of the hidden layer to prevent the vanishing gradient problem: the number of hidden layer nodes is selected according to the following empirical formula. Through the non-linear mapping ability of the MLP model, calibrate the Weibull distribution parameters, and use the physical and chemical correlation data as the validation set to optimize the model to prevent overfitting and improve the accuracy of the reliability function.

[0055] Step ⑥: Obtain the aging function expression and evaluate the aging state of the superconducting cable system

[0056] Through the calibrated Weibull distribution parameters, calculate the real-time reliability of the superconducting cable system and predict the aging degree evaluation index. Divide the 2000 sample data sets into a training set and a test set according to the ratio of 80%:20% for the training and evaluation of the model.

[0057] Among them, 1600 samples in the training set are used to train the model. Through the training set, the model learns the rules in the data and adjusts the parameters to minimize the loss function. 400 samples in the test set are used to evaluate the generalization ability of the model. The data in the test set does not participate in the learning of the model during the training process, so it can objectively reflect the performance of the model on unknown data.

[0058] Finally, the function of the superconducting cable aging state evaluation system based on mixed data is verified through 2,000 sample data sets. By comparing the true values and predicted values of the aging degree evaluation index, the performance of the training set is as follows Figure 4 shown. The model has a good fitting effect on the training set and can capture the patterns in the data well. The loss function value of the training set is low, indicating that the prediction error of the model on the training data is small. The performance of the test set is as follows Figure 5 shown. The model also performs well on the test set, indicating that the model has strong generalization ability. The loss function value of the test set is close to that of the training set, indicating that there is no obvious overfitting phenomenon in the model. The aging level is divided according to the aging degree evaluation index, and an aging state evaluation report of the superconducting cable system is generated, including the current reliability value, aging level, and maintenance suggestions

[0059] In summary, the system described in the present invention is applicable to the aging analysis of superconducting cable systems under multiple working conditions, including normal operation conditions, fault conditions, and extreme environmental conditions. By analyzing mixed data, the operating state of superconducting cables is evaluated, and aging analysis of superconducting cable systems under multiple working conditions is carried out to provide decision-making support for system maintenance

[0060] The present invention has been described in detail with reference to the embodiments accompanied by drawings. Those of ordinary skill in the art can make various variations of the present invention based on the above description. Therefore, some details in the embodiments should not constitute a limitation to the present invention, and the protection scope of the present invention will be defined by the scope of the appended claims

Claims

1. A superconducting cable aging state evaluation system based on hybrid data, characterized in that: The system includes a data analysis module, an aging assessment module, a physical-chemical correlation module, and a report generation module that are communicatively connected. Among them: The data analysis module includes a simulation sub-module, a data acquisition sub-module, and a data processing sub-module connected thereto. During the aging state assessment process carried out by the aging assessment module, the model parameters of the superconducting cable system are output and calibrated through the physical-chemical correlation model, and then an aging state assessment fitting function is output through the data processing sub-module, and an aging state assessment report is automatically generated by the report generation module.

2. The superconducting cable aging state evaluation system based on hybrid data according to claim 1, characterized in that, The simulation sub-module is used to simulate the operating state parameters of the superconducting cable system under fault conditions; the data acquisition sub-module includes a number of sensors for real-time monitoring of the operating state parameters and environmental state parameters of the superconducting cable system; the data processing sub-module is used to clean, standardize, and extract features from the collected data; The physical-chemical correlation module is used to calibrate the correlation data of the aging state of the superconducting cable system; the aging assessment module is used to calculate the aging function of the superconducting cable system and evaluate the real-time aging degree; the report generation module is used to generate an aging state assessment report of the superconducting cable system.

3. The superconducting cable aging state evaluation system based on hybrid data according to claim 2, characterized in that The simulation sub-module adopts a preset fault condition model to simulate the change trends of the current, voltage, and body temperature parameters of the superconducting cable system.

4. The superconducting cable aging state evaluation system based on hybrid data according to claim 2, wherein The number of sensors in the data acquisition sub-module are current sensors, voltage sensors, temperature sensors, and vacuum sensors, which are used to real-time monitor the current value, voltage value, body temperature value, environmental temperature value, and vacuum value of the superconducting cable system.

5. The superconducting cable aging state evaluation system based on hybrid data according to claim 2, characterized in that, The data processing sub-module removes abnormal data by adopting a data cleaning algorithm and unifies the data of different parameters to the same dimension by adopting a standardization algorithm.

6. The superconducting cable aging state evaluation system based on hybrid data according to claim 2, characterized in that, The physical-chemical correlation module adopts a physical-chemical analysis method to establish the correlation relationship between the operating state parameters and the aging state of the superconducting cable system and calibrate the correlation data.

7. The superconducting cable aging state evaluation system based on hybrid data according to claim 2, characterized in that The aging assessment module adopts a preset aging function, combines the real-time collected operating state parameters and environmental state parameters, calculates the real-time aging degree of the superconducting cable system, and divides the aging grades.

8. A method for assessing the aging state of a superconducting cable based on the system according to any one of claims 1 to 7, comprising the following steps: Step ①: Construct a multivariate Weibull distribution model F(X); Step ②: Obtain the joint probability density function f(x) by taking partial derivatives; Step ③: Organize the log-likelihood function log(f(x)); Step ④: Calculate the minimum value of the negative log-likelihood function to obtain the estimated values of the model parameters; Step ⑤: Fit the model parameters through a multi-layer perceptron neural network algorithm; 9. A method for evaluating the aging state of a superconducting cable according to claim 8, characterized in that: Step ⑥: Obtain the aging function expression and evaluate the aging state of the superconducting cable system. K characteristic parameters \(X(t)=[X_1(t),\ldots,X K (t)]\) follow a multivariate Weibull distribution. By constructing a degradation distribution function \(F(X)\) with \(K\) characteristic parameters, the reliability function of the superconducting cable system at time \(t\) is where η(t) = (η1(t), …, η K (t)) is the scale parameter; β(t) = (β1(t), …, β K (t)) is the shape parameter; θ(t) = (θ1(t), …, θ K (t)) is the correlation parameter; D f (t) = (D f1 (t), …, D fK (t)) is the aging assessment index; g(t) is the life parameter.

10. A method for evaluating the aging state of a superconducting cable according to claim 8, characterized in that: The implementation process of the said Step ① is as follows: The implementation process of the said Step ⑤ is as follows: Construct a multi-layer perceptron neural network model, including an input layer, a hidden layer, and an output layer. The model parameters of the superconducting cable system are input through the input layer, and the aging state assessment fitting function is output through the output layer. Use the backpropagation algorithm to train the neural network model and adjust the weights and bias terms.

Images