Detection method of glass fiber tape for superconducting magnet insulation

Abstract

The invention relates to the technical field of material performance detection, and discloses a method for detecting a glass fiber tape for superconducting magnet insulation, which comprises the following steps: acquiring a plurality of groups of glass fiber tape sample assemblies treated by different processes, each group of glass fiber tape sample assembly comprising two glass fiber tape samples; acquiring infrared spectrum data of one glass fiber ribbon sample in each group of glass fiber ribbon sample assembly; acquiring component and concentration information of another glass fiber belt sample in each group of glass fiber belt sample assembly, and acquiring mechanical property data; taking infrared spectrum data as input, taking component and concentration information as output, and performing training through an algorithm model so as to establish an infrared spectrum quantitative inversion model; and by taking the infrared spectrum data as input and the mechanical property data as output, training through an algorithm model so as to establish a mechanical property inversion model of the glass fiber belt. According to the invention, nondestructive testing can be carried out, the change of the molecular structure can be rapidly analyzed by using infrared spectrum, and the trend of mechanical properties can be predicted.

Description

Technical Field

[0001] The invention relates to the technical field of material performance detection, in particular to a detection method for a glass fiber tape for superconducting magnet insulation. Background Art

[0002] Fusion reactor superconducting magnets are developing towards high field strengths, high currents, and high stresses. Currently, the primary material used for high-field magnets is Nb3Sn (niobium tin) superconducting material, which can support higher magnetic field intensities and current densities. Glass fiber has excellent mechanical properties, electrical insulation properties, chemical stability, and resistance to corrosion and high temperatures. Therefore, glass fiber has been widely used for the insulation wrapping of Nb3Sn (niobium tin) superconducting magnets.

[0003] The widely used glass fiber tape materials, when used in superconducting magnet insulation, require vacuum heat treatment, carbon removal, and surface modification, resulting in significant variations in the mechanical properties of superconducting magnet insulation structures. Testing the internal structure and mechanical properties of superconducting magnet insulation structures involves fracture strength testing, scanning electron microscopy (SEM), thermogravimetry-differential scanning calorimetry (TG-DSC), interlaminar shear strength testing, and electrical insulation performance testing. These tests, which involve the mechanical, insulating, and thermodynamic properties of the insulation material, require the use of multiple different scientific research equipment and collaborative testing by multiple departments and technicians. Testing a batch of samples typically takes over a week, making it difficult to effectively support the production efficiency and quality testing requirements of superconducting magnet insulation structures. Furthermore, the testing process can cause physical damage to the glass fiber tape samples, which undoubtedly increases the cost of using the complex and expensive processing of the glass fiber tape samples, as well as samples that require subsequent testing. Summary of the Invention

[0004] The purpose of the present invention is to provide a method for detecting glass fiber tape for superconducting magnet insulation, so as to solve the problem that the current insulation testing method is difficult to non-destructively, quickly and accurately characterize the overall condition of the glass fiber tape.

[0005] The technical solution of the present invention is: A method for detecting a glass fiber tape for superconducting magnet insulation comprises the following steps: Obtain multiple sets of glass fiber tape sample assemblies that have been processed by different processes, each set of glass fiber tape sample assemblies containing two glass fiber tape samples; Obtain infrared spectrum data of one glass fiber tape sample in each set of glass fiber tape sample assemblies; Obtaining component and concentration information of another glass fiber tape sample in each group of glass fiber tape sample assemblies as first test data, and obtaining mechanical properties as second test data; The infrared spectrum data is used as input and the corresponding first test data is used as output. The infrared spectrum quantitative inversion model is established through training through the algorithm model. The trained infrared spectrum quantitative inversion model is used to perform qualitative and quantitative detection of surface attachments of unknown glass fiber tape samples. Taking infrared spectrum data as input and the corresponding second test data as output, an algorithm model is trained to establish an inversion model for the mechanical properties of glass fiber tapes. The trained inversion model for the mechanical properties of glass fiber tapes is used to test the mechanical properties of unknown glass fiber tape samples.

[0006] Preferably, as a further improvement of the present invention, after obtaining the infrared spectrum data, the infrared spectrum data is subjected to noise reduction processing, and the noise reduction processing steps include: Polynomial fitting method was used to eliminate the influence of background signals; The Savitzky-Golay smoothing method is used to reduce the noise in the spectrum.

[0007] Preferably, as a further improvement of the present invention, the algorithm model is a support vector machine.

[0008] Preferably, as a further improvement of the present invention, an infrared spectroscopy quantitative inversion model is used to perform qualitative and quantitative detection of surface attachments of unknown glass fiber tape samples, including characteristic peak identification and spectral matching and retrieval judgment. The characteristic peak identification judges the type of functional groups or compounds present in the sample by identifying the characteristic peaks of the infrared spectrum, thereby obtaining the changes in the organic matter components attached to the glass fiber tape through changes in the infrared spectrum. The spectral matching and retrieval judgment includes matching the measured spectrum with a known standard spectrum database, calculating the similarity between the measured spectrum and the spectrum in the database through the distance best similarity algorithm after spectral normalization, and determining the matched substance by analyzing the similarity index between the sample spectrum and the spectra of all samples in the spectral library.

[0009] Preferably, as a further improvement of the present invention, the infrared spectrum data of the glass fiber tape sample is obtained by measuring with a Fourier transform infrared spectrometer.

[0010] Preferably, as a further improvement of the present invention, the obtaining of the component and concentration information of the glass fiber tape sample is achieved by gas chromatography or X-ray diffraction.

[0011] Preferably, as a further improvement of the present invention, the mechanical properties of the glass fiber tape sample are obtained by using a stretching machine.

[0012] Compared with the prior art, the present invention has the following beneficial effects: 1. Infrared spectroscopy does not physically damage the fiberglass tape used for superconducting magnet insulation, allowing for the acquisition of internal structural information and mechanical properties without compromising sample integrity. This is crucial for fiberglass tape samples, which are complex and expensive to process, and for samples requiring subsequent testing.

[0013] 2. By detecting the infrared absorption peaks of different chemical bonds, such as silicon-oxygen bonds and carbon-hydrogen bonds, we can accurately identify the molecular vibration modes within the fiberglass tape and analyze the type, number, and spatial arrangement of the chemical bonds. This microscopic analysis is key to understanding the nature of the mechanical properties of fiberglass tape and helps us understand the root causes of material performance differences.

[0014] 3. It can reflect a variety of mechanical properties, such as tensile strength, bending properties and impact properties.

[0015] 4. Only a simple infrared spectrum test is needed to quickly predict the mechanical properties of glass fiber tapes, which greatly improves production efficiency and quality control level and provides a strong basis for product performance optimization.

[0016] 5. Compared to some traditional mechanical property testing methods, infrared spectroscopy is faster. It does not require complex sample preparation or long testing cycles. It can be quickly tested during the fiberglass tape processing process, providing timely feedback on product quality and facilitating timely adjustments to the processing process. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 The figure is a flow chart of a method for detecting a glass fiber tape for superconducting magnet insulation according to an embodiment of the present invention. DETAILED DESCRIPTION

[0018] The following is combined with Figure 1 , a detailed description of the specific embodiments of the present invention is provided. In the description of the invention, it should be understood that the terms "center," "upper," "lower," "front," "back," "left," "right," "vertical," "horizontal," "top," "bottom," "inside," "outside," and the like, indicating directions or positional relationships, are based on the directions or positional relationships shown in the accompanying drawings and are intended only to facilitate the description of the present invention and simplify the description. They do not indicate or imply that the devices or components referred to must have a specific direction, be constructed, or operate in a specific direction. Therefore, they should not be construed as limiting the present invention.

[0019] The terms "first" and "second" are used for descriptive purposes only and should not be understood to indicate or imply relative importance or implicitly specify the quantity of the technical features indicated. Therefore, features specified as "first" or "second" may explicitly or implicitly include one or more of such features; and in the description of the invention, unless otherwise specified, "plurality" means two or more.

[0020] Example like Figure 1 As shown, an embodiment of the present invention provides a method for detecting a glass fiber tape for superconducting magnet insulation, comprising the following steps: S1. Obtain multiple groups of glass fiber tape sample assemblies that have been processed by different processes, each group of glass fiber tape sample assemblies containing two glass fiber tape samples.

[0021] Among them, the acquisition process refers to using a shearing tool to cut a section from the fiberglass tape as a sample. During the cutting process, the tested part cannot be directly touched by hand. Different processes include different parameters such as heat treatment temperature, impregnation time, and impregnation type.

[0022] S2. Place one glass fiber tape sample from each group of glass fiber tape sample assemblies on the probe of a Fourier transform infrared spectrometer, and use the Fourier transform infrared spectrometer to detect the glass fiber tape sample to obtain infrared spectrum data of the glass fiber tape sample.

[0023] A Fourier transform infrared spectrometer primarily consists of a Michelson interferometer and a computer. The Michelson interferometer's primary function is to split light from a light source into two beams, creating a specific optical path difference, and then recombining them to produce interference. The resulting interference pattern function contains all the frequency and intensity information of the light source. By Fourier transforming the interference pattern function using a computer, the frequency distribution of the original light source's intensity can be calculated. While simple fiberglass tape is made of SiO2, its infrared spectrum is stable and pure. However, fiberglass tape used for superconducting magnet insulation requires a series of treatments, including the addition of substances such as palmitic acid, which results in the attachment of large amounts of organic matter to the tape. The content and type of organic matter reflect the insulating and electrical properties of the tape.

[0024] Considering the presence of background signal noise in the acquired infrared spectral data, background correction and smoothing are required. Because background signals may interfere with target spectral signals, the present invention estimates and deducts the background using methods such as polynomial fitting. The background spectrum is obtained by repeatedly measuring the infrared spectrum of a pure glass fiber tape under standard conditions. The detection position of the sample has a certain influence on its infrared spectrum. The present invention uses a polynomial fitting method to eliminate the influence of the background spectrum and achieve background correction. Smoothing is primarily intended to reduce noise in the spectrum. To improve detection speed and spectral processing stability, the present invention uses the Savitzky-Golay smoothing method, a polynomial fitting-based smoothing method that can better preserve the spectral shape while smoothing, particularly for spectral curves with a certain curvature.

[0025] S4. Use gas chromatography or X-ray diffraction to obtain the composition and concentration information of another glass fiber tape sample in each group of glass fiber tape sample components as the first test data, and use a tensile machine and obtain the mechanical properties of the glass fiber tape sample according to the national standard GB / T 7689.5-2013 as the second test data.

[0026] S5. Take the infrared spectrum data as input and the corresponding first test data as output, train the algorithm model to establish an infrared spectrum quantitative inversion model, and use the trained infrared spectrum quantitative inversion model to perform qualitative and quantitative detection of surface attachments on unknown glass fiber tape samples.

[0027] Among them, the qualitative analysis and detection of the surface attachments of unknown glass fiber tape samples using the infrared spectroscopy quantitative inversion model includes characteristic peak identification and spectrum matching and retrieval judgment. Characteristic peak identification determines the type of functional groups or compounds present in the sample by identifying the characteristic peaks of the infrared spectrum, thereby obtaining the changes in the organic matter components attached to the glass fiber tape through the changes in the infrared spectrum. For example, different substances have their own unique spectral characteristic peaks, and the carbonyl group (C = O) has a peak at 1700 cm -1 ~ 1750 cm -1 There are characteristic absorption peaks near 1710cm. By identifying these characteristic peaks, we can determine the possible functional groups or compound types in the sample. When analyzing an organic compound, if the peak at 1710cm -1 If a strong absorption peak is detected nearby, it can be inferred that the compound may contain a carbonyl structure. Changes in the infrared spectrum of the glass fiber tape can be used to determine changes in the organic matter components attached to the tape. During the spectral matching and search process, the measured spectrum is matched against a database of known standard spectra. The similarity between the measured spectrum and the spectra in the database is calculated using a distance-based optimal similarity algorithm after normalization of the spectra. The matching substance is determined by using a similarity index between the spectrum of the analyzed sample and the spectra of all samples in the spectral library.

[0028] During the quantitative analysis process, a calibration curve was first established. By measuring the infrared spectra of glass fiber tapes treated with a series of different parameters, the type and content of attached organic matter were determined using standard methods. A relationship curve was then established between the infrared spectral intensity of the glass fiber tape and the amount of attached matter on the glass fiber tape. The infrared spectral intensity includes the absorption peak area and peak height. After establishing the calibration curve, the spectral parameters of unknown glass fiber tapes were measured to obtain information on the type and concentration of attached organic matter. This information was then matched to the physical and chemical properties of the treated glass fiber tapes. Finally, based on the infrared spectral quantitative inversion model, quantitative detection of attached matter on the surface of the glass fiber tape samples was achieved.

[0029] S6. Using the infrared spectrum data as input and the corresponding second test data as output, the algorithm model is trained to establish an inversion model for the mechanical properties of the glass fiber tape, and the trained inversion model for the mechanical properties of the glass fiber tape is used to test the mechanical properties of the unknown glass fiber tape sample.

[0030] Since the mechanical properties of the glass fiber tape show a significant nonlinear relationship with its infrared spectral data, a support vector machine regression method is used to establish a quantitative relationship model between the infrared spectral characteristics and its mechanical performance parameters. The low-dimensional data is mapped to a high-dimensional space through a kernel function, so that the data can be linearly separable in the high-dimensional space. The present invention uses a radial basis kernel function\(K(x_i,x_j)=\exp(-\gamma||x_i - x_j||^2)\), where\(\gamma\) is the kernel parameter. By adjusting the kernel parameter and the penalty parameter\(C\), the structural risk is minimized and a nonlinear relationship model between the infrared spectral characteristics and the mechanical performance parameters is constructed. When optimizing the model, k-fold cross-validation is used to divide the data set into k subsets. The model is trained with k-1 subsets each time and tested with the remaining 1 subset, repeated k times, and the average evaluation index is taken as the model performance index to reduce the error caused by the data set division and improve the generalization ability of the model.

[0031] The principle behind developing an inverse model for the mechanical properties of fiberglass tapes is as follows: The tensile strength of fiberglass tapes is largely dependent on the orientation of their molecular chains. When the molecular chains are highly oriented along the tensile direction of the tape, they effectively transfer stress, thereby increasing the tensile strength. Changes in certain characteristic absorption peaks in the infrared spectrum can reflect changes in molecular chain orientation. The intensity of certain vibrational mode absorption peaks in the infrared spectrum, which are related to the molecular chain orientation, changes. By analyzing these changes, the effect of molecular chain orientation on tensile strength can be indirectly assessed. The bending properties of fiberglass tapes are closely related to intermolecular forces. Strong intermolecular forces prevent relative sliding of the molecular chains during bending, resulting in high material rigidity. Certain absorption peaks in the infrared spectrum can reflect the strength of intermolecular forces, such as hydrogen bonds and van der Waals forces. For example, the presence of hydrogen bonds can cause certain absorption peaks to shift or change in intensity. By studying these changes, we can gain a deeper understanding of the mechanisms by which intermolecular forces influence the bending properties of fiberglass tapes. The impact performance of fiberglass tapes depends on the strength of their internal chemical bonds and the integrity of their molecular structure. During impact, chemical bonds must be able to absorb and dissipate energy to prevent the material from breaking. The intensity of the characteristic absorption peak of chemical bonds in infrared spectroscopy is directly related to the strength of the chemical bonds. Strong chemical bonds have larger absorption peaks in infrared spectroscopy. By analyzing the intensity of these absorption peaks, the contribution of the chemical bonds within the glass fiber tape to the impact performance can be evaluated.

[0032] In summary, in the research and development of glass fiber tapes for superconducting magnet insulation, researchers can quickly use infrared spectroscopy analysis to preliminarily screen out material formulas with potentially excellent mechanical properties. For example, when trying a new glass fiber tape synthesis process or adding new additives, with the help of infrared spectroscopy, changes in the molecular structure can be quickly understood and the trend of mechanical properties can be predicted. Compared with the traditional method of preparing a large number of samples first and conducting tedious mechanical tests one by one, the research and development cycle is greatly shortened, and manpower and material costs are saved. During the glass fiber tape processing process, this method can be used to monitor the quality of the glass fiber tape in real time. Once an abnormality is found in the infrared spectral characteristics, it means that there may be a problem with the mechanical properties, which can be avoided during use to avoid causing more serious losses. From the perspective of product quality control, this method provides a comprehensive means of quality assessment. By continuously monitoring the molecular structure of the glass fiber tape, it is ensured that the tensile strength, bending properties and impact properties of each batch of products are stable within the standard range, thereby improving the reliability of the product.

[0033] The above disclosures are only several preferred specific embodiments of the present invention. However, the embodiments of the present invention are not limited thereto. Any changes that can be conceived by those skilled in the art should fall within the scope of protection of the present invention.

Claims

1. A method for detecting glass fiber tape for superconducting magnet insulation, characterized in that: The following steps are involved: Obtain multiple sets of glass fiber tape sample assemblies that have been processed by different processes, each set of glass fiber tape sample assemblies containing two glass fiber tape samples; Obtain infrared spectrum data of one glass fiber tape sample in each set of glass fiber tape sample assemblies; Obtaining component and concentration information of another glass fiber tape sample in each group of glass fiber tape sample assemblies as first test data, and obtaining mechanical properties as second test data; The infrared spectrum data is used as input and the corresponding first test data is used as output. The infrared spectrum quantitative inversion model is established through training through the algorithm model. The trained infrared spectrum quantitative inversion model is used to perform qualitative and quantitative detection of surface attachments of unknown glass fiber tape samples. Taking infrared spectrum data as input and the corresponding second test data as output, an algorithm model is trained to establish an inversion model for the mechanical properties of glass fiber tapes. The trained inversion model for the mechanical properties of glass fiber tapes is used to test the mechanical properties of unknown glass fiber tape samples.

2. The method for detecting glass fiber tape for superconducting magnet insulation according to claim 1, characterized in that: After obtaining the infrared spectrum data, the infrared spectrum data is subjected to noise reduction processing, wherein the noise reduction processing steps include: Polynomial fitting method was used to eliminate the influence of background signals; The Savitzky-Golay smoothing method is used to reduce the noise in the spectrum.

3. The method for detecting glass fiber tape for superconducting magnet insulation according to claim 1, wherein: The algorithm model is a support vector machine.

4. The method for detecting glass fiber tape for superconducting magnet insulation according to claim 1, wherein: An infrared spectroscopy quantitative inversion model is used to perform qualitative and quantitative detection of surface attachments on unknown glass fiber tape samples, including characteristic peak identification and spectrum matching and retrieval judgment. The characteristic peak identification determines the type of functional groups or compounds present in the sample by identifying characteristic peaks of the infrared spectrum, thereby obtaining changes in the organic matter components attached to the glass fiber tape through changes in the infrared spectrum. The spectrum matching and retrieval judgment includes matching the measured spectrum with a known standard spectrum database, calculating the similarity between the measured spectrum and the spectrum in the database through the distance best similarity algorithm after normalization of the spectrum graph, and determining the matched substance by analyzing the similarity index between the sample spectrum and the spectra of all samples in the spectrum library.

5. The method for detecting glass fiber tape for superconducting magnet insulation according to claim 1, characterized in that: The infrared spectrum data of the glass fiber tape sample is obtained by Fourier transform infrared spectrometer.

6. The method for detecting glass fiber tape for superconducting magnet insulation according to claim 1, characterized in that: The component and concentration information of the glass fiber tape sample is obtained by gas chromatography or X-ray diffraction.

7. The method for detecting glass fiber tape for superconducting magnet insulation according to claim 1, characterized in that: The mechanical properties of the glass fiber tape sample are obtained by using a tensile machine.

Images