Method and device for predicting service life of superconducting magnet

Abstract

FIELD: data processing technology.SUBSTANCE: method for predicting the service life of a superconducting magnet comprises the steps of collecting temperature data of the superconducting magnet and determining a first service life parameter of the superconducting magnet based on the temperature data; collecting magnetic field strength data of the superconducting magnet and determining a second service life parameter of the superconducting magnet based on the magnetic field strength data; collecting air pressure data of the superconducting magnet and determining a third service life parameter of the superconducting magnet based on the air pressure data; predicting based on the first service life parameter, the second service life parameter and the third service life parameter, so as to obtain the service life of the superconducting magnet.EFFECT: increase in the accuracy of predicting the service life of a superconducting magnet in a superconducting state.10 cl, 5 dwg, 1 tbl

Description

Cross-reference to a related application

[0001] This application claims priority to Chinese Patent Application No. 202311575649.1, entitled "METHOD AND APPARATUS FOR PREDICTING SERVICE LIFE OF SUPERCONDUCTING MAGNET", filed on November 23, 2023, in the National Intellectual Property Administration of China, which is incorporated herein by reference in its entirety.Technical field to which the invention pertains

[0002] The present disclosure relates to the field of data processing technology, and in particular, to a method and apparatus for predicting the service life of a superconducting magnet.Background of the art

[0003] A high-temperature superconducting electrodynamic suspension train requires a superconducting magnet to maintain a superconducting state and generate a strong electromagnetic field, ensuring the suspension, traction, and guidance of the train. The autonomous operating mode refers to an operating mode in which the power source is disconnected, vacuum pumping is stopped, and the refrigerator is deactivated during train operation. In the autonomous operating mode, energy consumption is significantly reduced and the operating safety of the train is improved. However, for the application of the autonomous operating mode, strict requirements are placed on predicting the service life of the superconducting magnet in the superconducting state. Summary of the invention

[0004] In view of the above problems, a method and device for predicting the service life of a superconducting magnet are provided according to the present disclosure in order to improve the accuracy of predicting the service life of a superconducting magnet for a high-temperature superconducting electrodynamic suspension train in an autonomous state.

[0005] In order to achieve the above-described object, the technical solutions are provided in the present disclosure as follows.

[0006] A method for predicting the service life of a superconducting magnet is provided. The method includes: obtaining temperature data of the superconducting magnet and determining a first service life parameter of the superconducting magnet based on the temperature data; obtaining magnetic field strength data of the superconducting magnet and determining a second service life parameter of the superconducting magnet based on the magnetic field strength data; obtaining air pressure data of the superconducting magnet and determining a third service life parameter of the superconducting magnet based on the air pressure data; and predicting the service life of the superconducting magnet based on the first service life parameter, the second service life parameter, and the third service life parameter.

[0007] In an embodiment of the invention, the method further includes: extracting temperature data corresponding to a superconducting magnet from test data of the superconducting magnet and creating a temperature data set, wherein the temperature data is obtained to represent temperature variations over time during the entire process from the beginning to the end of autonomous operation of the superconducting magnet; performing a sensitivity analysis for the temperature data in the temperature data set in order to determine an initial set of training samples of the temperature data; normalizing the temperature data in the initial set of training samples in order to obtain a target set of training samples; and performing training based on the target set of training samples in order to obtain a service life prediction model based on the temperature data.Determining a first service life parameter of a superconducting magnet based on temperature data includes: processing the temperature data by using a service life prediction model based on the temperature data to obtain the first service life parameter.

[0008] In an embodiment of the invention, performing a sensitivity analysis for temperature data in a temperature data set in order to determine an initial set of training samples of the temperature data includes: determining the influence parameters of temperature data obtained at different positions at one point in time in the temperature data set on the service life of the superconducting magnet; and determining a target temperature acquisition position based on the influence parameters and determining the temperature data obtained at the target temperature acquisition position as an initial set of training samples.

[0009] In an embodiment of the invention, the method further includes: extracting magnetic field strength data corresponding to a superconducting magnet from test data of the superconducting magnet and creating a magnetic field strength data set, wherein the magnetic field strength data are obtained to represent variations in the magnetic field strength over time during the entire process from the beginning to the end of autonomous operation of the superconducting magnet; performing a sensitivity analysis for the magnetic field strength data in the magnetic field strength data set to determine an initial set of training samples of the magnetic field strength data; normalizing the magnetic field strength data in the initial set of training samples to obtain a target set of training samples;and performing training based on the target set of training samples to obtain a lifetime prediction model based on the magnetic field strength data. Determining a second lifetime parameter of a superconducting magnet based on the magnetic field strength data includes: processing the magnetic field strength data by using the lifetime prediction model based on the magnetic field strength data to obtain the second lifetime parameter.

[0010] In an embodiment of the invention, the method further includes: extracting air pressure data corresponding to a superconducting magnet from test data of the superconducting magnet and creating a set of air pressure data, wherein the air pressure data is obtained to represent variations in air pressure over time during the entire process from the beginning to the end of autonomous operation of the superconducting magnet; performing a sensitivity analysis for the air pressure data in the air pressure data set to determine an initial set of training samples of the air pressure data; normalizing the air pressure data in the initial set of training samples to obtain a target set of training samples; and performing training based on the target set of training samples to obtain a service life prediction model based on the air pressure data.Determining the third parameter of the service life of a superconducting magnet based on the air pressure data includes: processing the air pressure data by using a service life prediction model based on the air pressure data to obtain the third parameter of the service life.

[0011] In an embodiment of the invention, predicting the service life of a superconducting magnet based on a first service life parameter, a second service life parameter, and a third service life parameter includes: determining the parameters of the degree of influence of temperature data, magnetic field strength data, and air pressure data on the service life of the superconducting magnet; determining a weighting coefficient based on the degree of influence parameters; and calculating the first service life parameter, the second service life parameter, and the third service life parameter by using the weighting coefficient in order to obtain the service life of the superconducting magnet.

[0012] A device for predicting the service life of a superconducting magnet is provided. The device comprises a first detection unit, a second detection unit, a third detection unit, and a prediction unit.

[0013] The first determining unit is configured to receive temperature data of the superconducting magnet and determine a first service life parameter of the superconducting magnet based on the temperature data.

[0014] The second determining unit is configured to receive magnetic field strength data of the superconducting magnet and determine a second service life parameter of the superconducting magnet based on the magnetic field strength data.

[0015] The third determining unit is configured to receive air pressure data of the superconducting magnet and determine a third service life parameter of the superconducting magnet based on the air pressure data.

[0016] The prediction unit is configured to predict the service life of the superconducting magnet based on the first service life parameter, the second service life parameter, and the third service life parameter.

[0017] In an embodiment of the invention, the device further includes: a first model creation unit. The first model creation unit is configured to: extract temperature data corresponding to the superconducting magnet from test data of the superconducting magnet and create a temperature data set, wherein the temperature data is obtained to represent temperature variations over time during the entire process from the beginning to the end of the autonomous operation of the superconducting magnet; perform a sensitivity analysis of the temperature data in the temperature data set in order to determine an initial set of training samples of the temperature data; normalize the temperature data in the initial set of training samples in order to obtain a target set of training samples; and perform training based on the target set of training samples in order to obtain a service life prediction model based on the temperature data.The first determining unit is further configured to process the temperature data by using a service life prediction model based on the temperature data in order to obtain the first service life parameter.

[0018] In an embodiment of the invention, the device further includes: a second model creation unit. The second model creation unit is configured to: extract magnetic field strength data corresponding to the superconducting magnet from test data of the superconducting magnet and create a magnetic field strength data set, wherein the magnetic field strength data are obtained to represent variations in the magnetic field strength over time during the entire process from the beginning to the end of autonomous operation of the superconducting magnet; perform a sensitivity analysis of the magnetic field strength data in the magnetic field strength data set to determine an initial set of training samples of the magnetic field strength data; normalize the magnetic field strength data in the initial set of training samples to obtain a target set of training samples;and perform training based on the target set of training samples in order to obtain a service life prediction model based on the magnetic field strength data. The second determining unit is further configured to process the magnetic field strength data by using the service life prediction model based on the magnetic field strength data in order to obtain the second service life parameter.

[0019] In an embodiment of the invention, the device further includes: a third model creation unit.The third model creation unit is configured to: extract air pressure data corresponding to the superconducting magnet from test data of the superconducting magnet and create a set of air pressure data, wherein the air pressure data is obtained to represent variations in air pressure over time during the entire process from the beginning to the end of autonomous operation of the superconducting magnet; perform a sensitivity analysis for the air pressure data in the air pressure data set in order to determine an initial set of training samples of the air pressure data; normalize the air pressure data in the initial set of training samples in order to obtain a target set of training samples; and perform training based on the target set of training samples in order to obtain a service life prediction model based on the air pressure data.The third determining unit is further configured to: process the air pressure data by using a service life prediction model based on the air pressure data in order to obtain a third service life parameter.

[0020] A method and device for predicting the service life of a superconducting magnet according to the present disclosure are provided. The method includes: obtaining temperature data of the superconducting magnet and determining a first service life parameter of the superconducting magnet based on the temperature data; obtaining magnetic field strength data of the superconducting magnet and determining a second service life parameter of the superconducting magnet based on the magnetic field strength data; obtaining air pressure data of the superconducting magnet and determining a third service life parameter of the superconducting magnet based on the air pressure data; and predicting the service life of the superconducting magnet based on the first service life parameter, the second service life parameter, and the third service life parameter.Compared with the prior art technology, a more accurate prediction result can be obtained in the present disclosure by obtaining data for predicting the service life of a superconducting magnet from various physical fields, and the method according to the present disclosure is more widely applicable to predicting the service life of a superconducting magnet for a railway train on a high-temperature superconducting electrodynamic suspension in an autonomous state. Brief description of the drawings.

[0021] To clearly illustrate the technical solutions in the embodiments of the present disclosure or the prior art technology, the following briefly describes the drawings, which are referred to in order to describe the embodiments of the invention or the prior art technology. Obviously, the drawings in the following description represent only some embodiments of the present disclosure, and for those skilled in the art, other drawings can be derived from these drawings without any creative efforts.

[0022] Fig. 1 is a flow chart of a method for predicting the service life of a superconducting magnet according to an embodiment of the present disclosure;

[0023] Fig. 2 is a schematic diagram showing the application of a temperature sensor according to an embodiment of the present disclosure;

[0024] Fig. 3 is a schematic diagram of a model-based life prediction according to an embodiment of the present disclosure;

[0025] Fig. 4 is a schematic diagram showing the processing of a POS algorithm according to an embodiment of the present disclosure; and

[0026] Fig. 5 is a schematic structural diagram of an apparatus for predicting the service life of a superconducting magnet according to an embodiment of the present disclosure. Detailed Description of the Invention

[0027] In the following, the technical solutions in the embodiments of the present disclosure are clearly and completely described in conjunction with the drawings in the embodiments of the present disclosure. It is obvious that the described embodiments of the invention represent only some, and not all, embodiments of the present disclosure. Based on the embodiments of the present disclosure, all other embodiments of the invention obtained by those skilled in the art without any creative efforts whatsoever fall within the scope of protection of the present disclosure.

[0028] Terms such as "first," "second," and the like, in the detailed description, claims, and drawings of this disclosure are intended to distinguish between different objects, rather than to describe a specific sequence. Furthermore, the terms "comprise," "include," and any variations thereof are intended to encompass a non-exclusive inclusion. For example, a process, method, system, product, or device that includes a sequence of steps or blocks is not limited to the listed steps or blocks, but may include steps or blocks that are not listed.

[0029] To simplify the description of embodiments of the present disclosure, relevant terms of the present disclosure are described as follows.

[0030] The superconducting magnet refers to an electromagnet with a coil made of a type II superconductor, which exhibits a high transition temperature and exceptionally high critical magnetic field when processed at ultra-low temperatures. The superconducting magnet also serves as a power supply component for a train with superconducting electrodynamic suspension.

[0031] Ultra-low temperature processing refers to cooling by using a liquid nitrogen device at a working temperature of -180°C, about -200°C.

[0032] Ultra-strong magnetic field refers to a magnetic field with a strength greater than 5 T (tesla), generated using superconducting technology. Also included are ultra-high-intensity magnetic fields generated using pulsed technology, hybrid magnetic technology, or ultra-high-power electromagnet technology.

[0033] High vacuum refers to the degree of vacuum below 1.333×10 -1 up to 1.333×10 -6 Pa.

[0034] The traction force (also known as the hanging force or guiding force) refers to the electromagnetic force generated by the repulsion between the superconducting magnet and the magnetic field of the ground traction coil or figure-eight coil.

[0035] The high-temperature superconducting electromagnetic suspension train is a new type of train car that combines high-temperature superconducting technology with magnetic suspension technology. The train car injects a large current and generates a strong magnetic field based on the zero-resistance properties of high-temperature superconducting materials. This contactless transportation is supported, guided, and propelled by magnetic forces, achieved through the interaction between the superconducting magnet on the train car and the magnetic field of the grounding coil.

[0036] Autonomous operation refers to the working mode in which the power source is disconnected, vacuum pumping is stopped, and the refrigerator is deactivated during the operation of the train.

[0037] Quenching: Superconductors have three fundamental characteristics: zero resistance, the Meissner effect, and the Josephson effect. Critical temperature, critical magnetic field, and critical current are three important parameters for superconductors. Quenching refers to the process in which a superconducting magnet undergoes a phase transition and returns to a normal conductor when any of the parameters (temperature, magnetic field, and current) exceeds its critical value. When the operating temperature of a superconductor remains constant, the superconductor can stably return to the superconducting state only if the current flowing through it drops below the recovery current. The energy released from the magnet during quenching can cause a rapid increase in the local temperature of the magnet.Excessive temperature increases can damage the internal structure of the superconductor or even cause magnet burnout. Post-quench magnet safety analysis requires solving the quench transient, which includes current attenuation, temperature variations, and magnet voltage variations. This allows for an assessment of the thermal shock and voltage surge on the magnet during quenching. This allows the magnet design to be evaluated for magnetic field strength and uniformity, providing the basis for design considerations and the magnet's manufacturing process.

[0038] Sensitivity analysis refers to a method for studying and analyzing the sensitivity of the state or output of a system (or model) to variations in system parameters or environmental conditions. In optimization methods, sensitivity analysis is often used to study the stability of an optimal solution when the input data is inaccurate or changed. Furthermore, sensitivity analysis can be used to determine which parameters have a significant influence on the system or model.

[0039] Multivariate Regression Learning Predictive PSO-CNN Model: PSO refers to particle swarm optimization, which is a population-based stochastic optimization technology. The particle swarm-based algorithm simulates the swarming behavior of insects, animals, birds, and fish. These swarms search for food cooperatively, and each swarm member constantly adjusts its search mode by learning from its own experience and the experiences of other members. The multivariate Regression Learning Predictive PSO-CNN model constructs a regression model for training and analysis from data through an algorithm using a convolutional neural network based on the principle of particle swarm optimization (PSO-CNN).

[0040] The method for predicting the service life of a superconducting magnet according to an embodiment of the present disclosure is mainly applied to a high-temperature superconducting electrodynamic suspension (HTSS) train. In order to maintain the superconducting state of the superconducting magnet in the HTSS train, conditions of ultra-low temperature, strong magnetic field, and high vacuum are required. Therefore, parameters for predicting the service life of the superconducting magnet, which are obtained from various physical fields, are required to achieve accurate prediction of the service life. Refer to FIG. 1, which is a flowchart of the method for predicting the service life of a superconducting magnet according to an embodiment of the present disclosure. The method may include the following steps S101-S104.

[0041] In step S101, the temperature data of the superconducting magnet is obtained, and the first service life parameter of the superconducting magnet is determined based on the temperature data.

[0042] The temperature data includes the temperature of the superconducting magnet coil, the temperature of the cooling structure, and the temperature of the nitrogen fixing cavity. The temperature data can be obtained using a temperature sensor (as shown in Fig. 2). Specifically, the temperature data at the current time can be obtained. Thereafter, the service life of the superconducting magnet is predicted based on the obtained temperature data to obtain a first service life parameter. The first service life parameter represents the applicable service life of the superconducting magnet, predicted based on the currently measured temperature data. In an embodiment of the invention, the prediction of the first service life parameter of the superconducting magnet can be performed using a pre-established predictive model based on the temperature data. The predictive model can be obtained through training based on existing data of the superconducting magnet.

[0043] In step S102, the magnetic field strength data of the superconducting magnet is obtained, and the second service life parameter of the superconducting magnet is determined based on the magnetic field strength data.

[0044] Magnetic field strength data can be obtained using a magnetic field sensor mounted on an external cryostat at various positions of the superconducting magnet at different times. The service life of the superconducting magnet is then predicted based on the obtained magnetic field strength data to obtain a second service life parameter. The second service life parameter represents the applicable service life of the superconducting magnet, predicted based on the currently measured magnetic field strength data. In an embodiment of the invention, the second service life parameter of the superconducting magnet can be predicted using a pre-established predictive model based on the magnetic field strength data. The predictive model can be obtained through training based on existing magnetic field strength data of the superconducting magnet.

[0045] In step S103, the air pressure data of the superconducting magnet is obtained, and the third service life parameter of the superconducting magnet is determined based on the air pressure data.

[0046] Air pressure data is obtained from the air pressure in the vacuum pumping region between the external cryostat and the nitrogen fixation cavity. The service life of the superconducting magnet is then predicted based on the obtained air pressure data to obtain a third service life parameter. This third service life parameter represents the applicable service life of the superconducting magnet, predicted based on currently measured air pressure data. In an embodiment, the third service life parameter of the superconducting magnet can be predicted using a pre-established predictive model based on the air pressure data. The predictive model can be obtained through training based on existing air pressure data of the superconducting magnet.

[0047] In step S104, the service life of the superconducting magnet is predicted based on the first service life parameter, the second service life parameter, and the third service life parameter.

[0048] The service life of a superconducting magnet can be predicted based on the respective influences of temperature data (corresponding to the first service life parameter), magnetic field strength data (corresponding to the second service life parameter), and air pressure data (corresponding to the third service life parameter) on the service life. For example, the service life of a superconducting magnet is obtained through a weighted calculation based on empirical data.In an embodiment of the invention, predicting the service life of a superconducting magnet based on a first service life parameter, a second service life parameter, and a third service life parameter includes: determining the parameters of the degree of influence of temperature data, magnetic field strength data, and air pressure data on the service life of the superconducting magnet; determining a weighting factor based on the degree of influence parameters; and calculating the first service life parameter, the second service life parameter, and the third service life parameter by using the weighting factor in order to obtain the service life of the superconducting magnet.

[0049] For example, the first parameter of service life is represented as L T , the second service life parameter is represented as L M , the third service life parameter is represented as L A, and each of them is empirically multiplied by a weighting factor to obtain the predicted service life, which can be expressed as L=0.5L T +0.25L M +0.25L A .

[0050] A method for predicting the service life of a superconducting magnet according to an embodiment of the present disclosure is provided. The method includes: obtaining temperature data of the superconducting magnet and determining a first service life parameter of the superconducting magnet based on the temperature data; obtaining magnetic field strength data of the superconducting magnet and determining a second service life parameter of the superconducting magnet based on the magnetic field strength data; obtaining air pressure data of the superconducting magnet and determining a third service life parameter of the superconducting magnet based on the air pressure data; and predicting the service life of the superconducting magnet based on the first service life parameter, the second service life parameter, and the third service life parameter.In the present disclosure, by obtaining parameters for predicting the service life of a superconducting magnet from various physical fields, an accurate prediction result can be obtained, and the method is applicable to predicting the service life of a superconducting magnet for a railway train on a high-temperature superconducting electrodynamic suspension in an autonomous state.

[0051] The process of predicting the service life based on data from various physical fields is described in detail as follows. Referring to Fig. 3, which is a schematic diagram of a method for predicting the service life of a superconducting magnet in a high-temperature superconducting electrodynamic suspension train according to an embodiment of the present disclosure. In this method, the service life L T service life is predicted through a service life prediction model created based on a temperature data set, the life time L Mservice life is predicted through a service life prediction model created based on a data set of magnetic field strength, life L A The service life is predicted using a service life prediction model created based on an air pressure dataset, and the service life is predicted after weighting factors are determined based on empirical data in order to provide guidance for testing and subsequent engineering applications. The process of creating service life prediction models based on various types of datasets in an embodiment of the present disclosure is described in detail as follows.

[0052] In an embodiment of the invention, the method further includes: extracting temperature data corresponding to a superconducting magnet from test data of the superconducting magnet and creating a temperature data set, wherein the temperature data is obtained to represent temperature variations over time during the entire process from the beginning to the end of autonomous operation of the superconducting magnet; performing a sensitivity analysis for the temperature data in the temperature data set in order to determine an initial set of training samples of the temperature data; normalizing the temperature data in the initial set of training samples in order to obtain a target set of training samples; and performing training based on the target set of training samples in order to obtain a service life prediction model based on the temperature data.Determining a first service life parameter of a superconducting magnet based on temperature data includes: processing the temperature data by using a service life prediction model based on the temperature data to obtain the first service life parameter.

[0053] Additionally, performing a sensitivity analysis for the temperature data in the temperature data set in order to determine an initial set of training samples of the temperature data includes: determining the influence parameters of the temperature data obtained at different positions at one point in time in the temperature data set on the service life of the superconducting magnet; determining a target temperature acquisition position based on the influence parameters; and determining the temperature data obtained at the target temperature acquisition position as the initial set of training samples.

[0054] For example, based on the previous test result of a superconducting magnet, the temperature data related to the superconducting magnet is extracted to create a temperature data set such as Ti{t1, t2, ..., tn}, i=1, 2, ..., n. After that, a sensitivity analysis is performed on the data in the temperature data set, and the temperature data with the highest correlation is determined as the initial set of training samples. For example, four sets of temperature data with the highest correlation can be selected, which are represented as Ti{t1, t2, ..., tn}, i=1, 2, ..., 4. The data in the initial set of training samples can be normalized as follows:

[0055] wherex Rmax represents the maximum value in the temperature data, i x Rminrepresents the minimum value in the temperature data. The target set of training samples is obtained after normalization, for example, Ttesti{t1, t2, ..., tn}, i=1, 2, ..., 4. The data set is randomly divided into two mutually exclusive sets, which are a training set (70%) and a validation set (30%). In an embodiment of the present disclosure, a multivariate regression training predictive PSO-CNN model is used to train the temperature data with the highest correlation to obtain a service life prediction model. In addition, other model structures can be used for training, and the model structure is not limited in the embodiments of the present disclosure.

[0056] The PSO algorithm is an optimization algorithm developed by studying the foraging behavior of birds. The fundamental concept of the PSO algorithm comes from the theory of artificial life and evolutionary computation, finding an optimal solution through cooperation and information exchange between individual elements in a population. Before optimizing the CNN parameters using the PSO algorithm, the hyperparameters to be optimized are first determined. Parameter determination is crucial for the performance of the model. For optimization, two crucial parameters are determined in the CNN model: the batch size (λ) and the number of epochs (μ). Using these two parameters as the objects of optimization for the PSO algorithm, the process of determining the CNN hyperparameters based on the PSO algorithm is shown in Fig. 4.

[0057] In a particle-based algorithm, five key parameters must be defined: the inertia coefficient, the number of particle features, the maximum number of iterations, the individual cognitive learning coefficient, and the social cognitive learning coefficient. The specific settings of these parameters are shown in the table below: Parameter type Parameter value Coefficient of inertia 0,5 Number of particle features 5 Maximum number of iterations 5 Individual cognitive learning quotient 2 Social cognitive learning coefficient 2

[0058] A CNN model is constructed with one input layer, four convolutional layers, two pooling layers, and one output layer. The output layer includes a single neuron representing the predicted service life. The model's accuracy is validated using temperature data obtained from previous tests. The predicted service life error is L T service is managed within 95% for qualification, otherwise iterative training is performed again.

[0059] In an embodiment of the invention, the method further includes: extracting magnetic field strength data corresponding to a superconducting magnet from test data of the superconducting magnet and creating a magnetic field strength data set, wherein the magnetic field strength data are obtained to represent variations in the magnetic field strength over time during the entire process from the beginning to the end of autonomous operation of the superconducting magnet; performing a sensitivity analysis for the magnetic field strength data in the magnetic field strength data set to determine an initial set of training samples of the magnetic field strength data; normalizing the magnetic field strength data in the initial set of training samples to obtain a target set of training samples;and performing training based on the target set of training samples to obtain a lifetime prediction model based on the magnetic field strength data. Determining a second lifetime parameter of a superconducting magnet based on the magnetic field strength data includes: processing the magnetic field strength data by using the lifetime prediction model based on the magnetic field strength data to obtain the second lifetime parameter.

[0060] For example, based on the previous test result, the magnetic field strength data of the superconducting magnet is extracted to create a magnetic field strength data set Mi{m1, m2, ..., mn}, i=1,2. Then, through sensitivity analysis, the magnetic field data at several positions (e.g., four positions) with the greatest influence at different times are determined as the initial set of training samples. The training samples are normalized as follows: In this case x Rmax represents the maximum value in the data, i Rmin represents the minimum value in the data. After normalization, the target set of training samples is obtained as Mtesti{m1, m2, ..., mn}, i=1,2.

[0061] The magnetic field strength data is trained using a regression-based learning predictive PSO-CNN model to obtain a service life prediction model. The accuracy of the model is validated using magnetic field strength data obtained from previous tests. The predicted service life error is L M service is managed within 95% for qualification, otherwise iterative training is performed again.

[0062] Similarly, in an embodiment of the present disclosure, the method further includes: extracting air pressure data corresponding to a superconducting magnet from test data of the superconducting magnet and creating a set of air pressure data, wherein the air pressure data is obtained to represent variations in air pressure over time during the entire process from the beginning to the end of autonomous operation of the superconducting magnet; performing a sensitivity analysis for the air pressure data in the air pressure data set to determine an initial set of training samples of the air pressure data; normalizing the air pressure data in the initial set of training samples to obtain a target set of training samples; and performing training based on the target set of training samples to obtain a service life prediction model based on the air pressure data.Determining the third parameter of the service life of a superconducting magnet based on the air pressure data includes: processing the air pressure data by using a service life prediction model based on the air pressure data to obtain the third parameter of the service life.

[0063] For example, based on the previous test result, the air pressure data of the superconducting magnet is extracted to create an air pressure data set Ai{a1, a2, ..., an}, i=1,2. Afterwards, sensitivity analysis and normalization are performed on the air pressure data to obtain the target training set. The normalization is performed as follows: In this case x Rmax represents the maximum value in the data, i Rmin represents the minimum value in the data. After normalization, the training set is obtained as Atesti{a1, a2, ..., an}, i=1,2.

[0064] The air pressure data is trained using a regression-based PSO-CNN predictive learning model to obtain a service life prediction model. The accuracy of the model is validated using air pressure data obtained from previous tests. The predicted service life error is L A service is managed within 95% for qualification, otherwise iterative training is performed again.

[0065] Terms L T , L M and L A services are predicted based on measured data of temperature, magnetic field strength and air pressure, and they are empirically multiplied by the appropriate weighting factors to obtain the predicted service life L=0.5L T +0.25L M +0.25L A , which provides guidance for testing and subsequent engineering applications.

[0066] When predicting the service life of a superconducting magnet in a high-temperature superconducting electrodynamic suspension (HTSS) train, conventional methods typically rely on engineers' experience in predicting the service life based on the temperature rise during testing. This does not guarantee the accuracy of the prediction and does not allow the service life of the superconducting magnet to be predicted before testing. If the predicted service life is underestimated, the capabilities of the superconducting magnet cannot be fully utilized. If the predicted service life is excessive, quenching may occur during operation due to the superconducting magnet. Therefore, a method for predicting the service life of a superconducting magnet in a high-temperature superconducting electrodynamic suspension (HTSS) train is provided according to the present disclosure.In this method, the parameters for predicting the service life of a superconducting magnet are obtained from various physical fields to obtain an accurately predicted service life, filling the research gap in methods for predicting the service life of a superconducting magnet for a high-temperature superconducting electrodynamic suspension (HTSC) train in an autonomous state.

[0067] Additionally, a device for predicting the service life of a superconducting magnet according to an embodiment of the present disclosure is provided. Referring to Fig. 5, the device includes a first determining unit 201, a second determining unit 202, a third determining unit 203, and a predicting unit 204.

[0068] The first determining unit 201 is configured to receive temperature data of the superconducting magnet and determine a first service life parameter of the superconducting magnet based on the temperature data.

[0069] The second determining unit 202 is configured to receive data of the magnetic field strength of the superconducting magnet and determine a second parameter of the service life of the superconducting magnet based on the data of the magnetic field strength.

[0070] The third determining unit 203 is configured to receive air pressure data of the superconducting magnet and determine a third service life parameter of the superconducting magnet based on the air pressure data.

[0071] The prediction unit 204 is configured to predict the service life of the superconducting magnet based on the first service life parameter, the second service life parameter, and the third service life parameter.

[0072] In an embodiment of the invention, the device further includes a first model creation unit. The first model creation unit is configured to: extract temperature data corresponding to the superconducting magnet from test data of the superconducting magnet and create a temperature data set, wherein the temperature data is obtained to represent temperature variations over time during the entire process from the beginning to the end of the autonomous operation of the superconducting magnet; perform a sensitivity analysis of the temperature data in the temperature data set to determine an initial set of training samples of the temperature data; normalize the temperature data in the initial set of training samples to obtain a target set of training samples; and perform training based on the target set of training samples to obtain a service life prediction model based on the temperature data.The first determining unit is further configured to process the temperature data by using a service life prediction model based on the temperature data in order to obtain the first service life parameter.

[0073] Additionally, performing a sensitivity analysis for the temperature data in the temperature data set in order to determine an initial set of training samples of the temperature data includes: determining the influence parameters of the temperature data obtained at different positions at one point in time in the temperature data set on the service life of the superconducting magnet; and determining a target temperature acquisition position based on the influence parameters and determining the temperature data obtained at the target temperature acquisition position as the initial set of training samples.

[0074] In an embodiment of the invention, the device further includes a second model creation unit. The second model creation unit is configured to: extract magnetic field strength data corresponding to the superconducting magnet from test data of the superconducting magnet and create a magnetic field strength data set, wherein the magnetic field strength data are obtained to represent variations in the magnetic field strength over time during the entire process from the beginning to the end of autonomous operation of the superconducting magnet; perform a sensitivity analysis of the magnetic field strength data in the magnetic field strength data set to determine an initial set of training samples of the magnetic field strength data; normalize the magnetic field strength data in the initial set of training samples to obtain a target set of training samples;and perform training based on the target set of training samples in order to obtain a service life prediction model based on the magnetic field strength data. The second determining unit is further configured to process the magnetic field strength data by using the service life prediction model based on the magnetic field strength data in order to obtain the second service life parameter.

[0075] In an embodiment of the invention, the device further includes a third model creation unit.The third model creation unit is configured to: extract air pressure data corresponding to the superconducting magnet from test data of the superconducting magnet and create a set of air pressure data, wherein the air pressure data is obtained to represent variations in air pressure over time during the entire process from the beginning to the end of autonomous operation of the superconducting magnet; perform a sensitivity analysis for the air pressure data in the air pressure data set in order to determine an initial set of training samples of the air pressure data; normalize the air pressure data in the initial set of training samples in order to obtain a target set of training samples; and perform training based on the target set of training samples in order to obtain a service life prediction model based on the air pressure data.The third determining unit is further configured to: process the air pressure data by using a service life prediction model based on the air pressure data in order to obtain a third service life parameter.

[0076] In an embodiment of the invention, the predicting unit is further configured to: determine the parameters of the degree of influence of the temperature data, the magnetic field strength data and the air pressure data on the service life of the superconducting magnet; determine a weighting coefficient based on the parameters of the degree of influence; and calculate a first service life parameter, a second service life parameter and a third service life parameter by using the weighting coefficient in order to obtain the service life of the superconducting magnet.

[0077] Based on the above embodiments of the invention, a machine-readable storage medium according to an embodiment of the present disclosure is provided. The machine-readable storage medium stores one or more programs. The one or more programs are executed by one or more processors to implement the method for predicting the service life of a superconducting magnet, as described above.

[0078] An electronic device according to an embodiment of the present disclosure is further provided. The electronic device includes a memory device, a processor, and a computer program stored in the memory device and executed by the processor. The computer program, when executed by the processor, instructs the processor to implement a method for predicting the service life of a superconducting magnet, as described above.

[0079] Only specific embodiments of the present disclosure have been described above, and the protection scope of the present disclosure is not limited thereto. Any modifications or substitutions that can be readily envisaged by those skilled in the art within the technical scope disclosed in the present disclosure shall be encompassed by the scope of the present disclosure. Therefore, the protection scope of the present disclosure shall be subject to the protection scope defined by the claims.

[0080] The embodiments in the description of the invention are presented sequentially, and each embodiment of the invention emphasizes the difference with respect to other embodiments of the invention. Identical and similar parts in various embodiments of the invention may be related to each other. Since the device disclosed in the embodiments of the invention corresponds to the method disclosed in the embodiments of the invention, the description of the device is simple, and reference should be made to the relevant part of the methods.

[0081] The foregoing description of the disclosed embodiments of the invention enables those skilled in the art to make or use the present disclosure. Various modifications to the embodiments of the invention will be obvious to those skilled in the art. The general principle defined herein can be implemented in other embodiments of the invention without departing from the spirit and scope of the present disclosure. Therefore, the present disclosure should not be limited to the embodiments of the invention disclosed herein, but has the widest scope consistent with the concept and novel features disclosed herein.

Claims

1. A method for predicting the service life of a superconducting magnet, comprising the steps of: obtaining the temperature data of the superconducting magnet and determining the first service life parameter of the superconducting magnet based on the temperature data; obtaining the magnetic field strength data of the superconducting magnet and determining the second service life parameter of the superconducting magnet based on the magnetic field strength data; obtaining air pressure data in the vacuum pumping region between the external cryostat and the nitrogen fixation cavity in the superconducting magnet and determining the third life parameter of the superconducting magnet based on the air pressure data; and predict the service life of a superconducting magnet based on the first service life parameter, the second service life parameter and the third service life parameter by means of the following equation L=0.5L T +0.25L M +0.25L A, where L is the predicted service life, L T represents the first parameter of service life, L M represents the second parameter of service life and L A represents the third parameter of service life.

2. A method for predicting the service life of a superconducting magnet according to claim 1, further comprising the steps of: extracting temperature data corresponding to the superconducting magnet from the test data of the superconducting magnet and creating a temperature data set, wherein the temperature data is obtained to represent the variations of the temperature over time during the entire process from the beginning to the end of the autonomous operation of the superconducting magnet; performing a sensitivity analysis on the temperature data in the temperature data set to determine an initial set of training temperature data samples; normalize the temperature data in the initial set of training samples to obtain the target set of training samples; and perform training on a target set of training samples to obtain a service life prediction model based on temperature data, wherein determining the first service life parameter of the superconducting magnet based on the temperature data comprises the step of: processing the temperature data by using a service life prediction model based on the temperature data in order to obtain the first service life parameter.

3. The method for predicting the service life of a superconducting magnet according to claim 2, wherein performing a sensitivity analysis for temperature data in a temperature data set in order to determine an initial set of training samples of temperature data comprises the steps of: determine the parameters of the influence of temperature data obtained at different positions at one point in time in the temperature data set on the service life of the superconducting magnet; and determining a target temperature acquisition position based on the influence parameters and determining temperature data acquired at the target temperature acquisition position as an initial set of training samples.

4. The method for predicting the service life of a superconducting magnet according to claim 1, further comprising the steps of: extracting magnetic field strength data corresponding to the superconducting magnet from the test data of the superconducting magnet and creating a set of magnetic field strength data, wherein the magnetic field strength data is obtained to represent variations in the magnetic field strength over time during the entire process from the beginning to the end of autonomous operation of the superconducting magnet; performing a sensitivity analysis on the magnetic field strength data in a magnetic field strength data set to determine an initial set of training samples of magnetic field strength data; normalize the magnetic field strength data in the initial set of training samples to obtain the target set of training samples; and performing training based on a target set of training samples in order to obtain a lifetime prediction model based on magnetic field strength data, wherein determining a second lifetime parameter of a superconducting magnet based on magnetic field strength data comprises a step in which: processing magnetic field strength data by using the lifetime prediction model based on magnetic field strength data in order to obtain the second lifetime parameter.

5. The method for predicting the service life of a superconducting magnet according to claim 1, further comprising the steps of: extracting air pressure data corresponding to the superconducting magnet from the test data of the superconducting magnet and creating a set of air pressure data, wherein the air pressure data is obtained to represent variations in air pressure over time during the entire process from the beginning to the end of autonomous operation of the superconducting magnet; performing a sensitivity analysis on the air pressure data in the air pressure data set to determine an initial set of training samples for the air pressure data; normalize the air pressure data in the initial set of training samples to obtain the target set of training samples; and perform training on a target set of training samples to obtain a service life prediction model based on air pressure data, wherein determining the third service life parameter of the superconducting magnet based on the air pressure data comprises the step of: processing the air pressure data by using a service life prediction model based on the air pressure data in order to obtain the third service life parameter.

6. The method for predicting the service life of a superconducting magnet according to claim 1, wherein predicting the service life of the superconducting magnet based on the first service life parameter, the second service life parameter and the third service life parameter comprises the steps of: determine the parameters of the degree of influence of temperature data, magnetic field strength data and air pressure data on the service life of a superconducting magnet; determine the weighting factor based on the parameters of the degree of influence; and calculate the first service life parameter, the second service life parameter and the third service life parameter by using the weighting coefficient to obtain the service life of the superconducting magnet.

7. A device for predicting the service life of a superconducting magnet, comprising: a first determining unit configured to receive temperature data of the superconducting magnet and to determine a first service life parameter of the superconducting magnet based on the temperature data; a second determining unit configured to receive magnetic field strength data of the superconducting magnet and to determine a second service life parameter of the superconducting magnet based on the magnetic field strength data; a third determining unit configured to receive air pressure data in the vacuum pumping region between the external cryostat and the nitrogen fixation cavity in the superconducting magnet and to determine a third service life parameter of the superconducting magnet based on the air pressure data; and a prediction unit configured to predict the service life of a superconducting magnet based on a first service life parameter, a second service life parameter and a third service life parameter by means of the following equation L=0.5L T +0.25L M +0.25L A , where L is the predicted service life, L T represents the first parameter of service life, L M represents the second parameter of service life and L A represents the third parameter of service life.

8. A device for predicting the service life of a superconducting magnet according to claim 7, further comprising a first model creation unit configured to: extract temperature data corresponding to the superconducting magnet from test data of the superconducting magnet and create a temperature data set, wherein the temperature data is obtained to represent variations in temperature over time during the entire process from the beginning to the end of autonomous operation of the superconducting magnet; perform sensitivity analysis of temperature data in a temperature data set to determine an initial set of training temperature data samples; normalize the temperature data in the initial set of training samples to obtain the target set of training samples; and perform training on a target set of training samples to obtain a service life prediction model based on temperature data, wherein the first determining unit is further configured to process the temperature data by using a service life prediction model based on the temperature data in order to obtain the first service life parameter.

9. A device for predicting the service life of a superconducting magnet according to claim 7, further comprising a second model creation unit configured to: extract magnetic field strength data corresponding to the superconducting magnet from test data of the superconducting magnet and create a magnetic field strength data set, wherein the magnetic field strength data is obtained to represent variations in the magnetic field strength over time during the entire process from the beginning to the end of autonomous operation of the superconducting magnet; perform a sensitivity analysis of the magnetic field strength data in the magnetic field strength data set in order to determine the initial set of training samples of magnetic field strength data; normalize the magnetic field strength data in the initial set of training samples to obtain the target set of training samples; and perform training on a target set of training samples to obtain a lifetime prediction model based on magnetic field strength data, wherein the second determining unit is further configured to process the magnetic field strength data by using a service life prediction model based on the magnetic field strength data in order to obtain the second service life parameter.

10. A device for predicting the service life of a superconducting magnet according to claim 7, further comprising a third model creation unit configured to: extract air pressure data corresponding to the superconducting magnet from the test data of the superconducting magnet and create a set of air pressure data, wherein the air pressure data is obtained to represent variations in the air pressure over time during the entire process from the beginning to the end of the autonomous operation of the superconducting magnet; perform a sensitivity analysis on the air pressure data in the air pressure data set to determine an initial set of training air pressure data samples; normalize the air pressure data in the initial set of training samples to obtain the target set of training samples; and perform training on a target set of training samples to obtain a service life prediction model based on air pressure data, wherein the third determining unit is further configured to process the air pressure data by using a service life prediction model based on the air pressure data in order to obtain a third service life parameter.

Images