RU2816673C1 - Method of detecting latent defects in composite materials by standing waves - Google Patents

Abstract

FIELD: measurement.

SUBSTANCE: use for detection of latent defects of composite materials. Essence of the invention consists in the fact that on the analysed article made of composite material, noise records are recorded with a sampling frequency of 20 Hz, noise recording is divided into blocks, which consist of a certain number of readings, then each block is divided into fragments with duration of 65536 counts and the amplitude spectra of these fragments are calculated and accumulated, then for each observation point the amplitude spectra of all fragments are averaged, the presence of standing waves is established, wherein variation of compression-expansion modes indicates presence of defects in composite material.

EFFECT: enabling detection of hidden defects in composite materials with high accuracy and speed of measurements without preliminary treatment of the surface of the analysed material and without using artificial seismoacoustic noise.

3 cl, 4 dwg

Description

The invention relates to the field of geophysical methods for qualitative assessment, diagnosis and localization of hidden defects in composite materials.

Currently, composite materials play an important role in various industries. Due to their unique properties, such as strength, durability, resistance to corrosion and environmental influences, they are widely used in the production of all types of transport, construction, biomedical products, etc. However, despite all the advantages, composite materials are susceptible to defects that can significantly reduce their efficiency and reliability. Examples of defects can be cracks, delaminations, foreign inclusions, cavities, etc. Therefore, an urgent task is to develop methods for detecting such defects, as well as studying their effect on the properties of the material.

State of the art.

At the stage of testing strength characteristics under cyclically repeated loads on products made of composite materials, there is a problem of heterogeneity of the results obtained: a number of products break, withstanding a load of ~ 1,000,000 repetitions, a number - an order of magnitude less. Since each specific part is manufactured in small batches from layers containing about 50 products, such significant differences in the load-bearing indicators at the strength testing stage may indicate problems existing in the technological chain and leading to such inhomogeneities in the results.

This problem can be voiced as a problem of diagnosing products at all stages of their manufacture in order to identify and localize possible defects that are visually indistinguishable to the naked eye, in order to ensure the requirements for the quality and safety of their operation.

To detect defects in composite materials, it is possible to use various non-destructive testing methods: acoustic, optical, thermal, radiation, magnetic, etc.

The optical-acoustic method of non-destructive testing is based on the principle of converting acoustic energy into optical energy and vice versa. It includes three main stages: generation of acoustic waves by a sound source that penetrate into the object under study, conversion of acoustic energy into thermal energy, which leads to a local change in temperature, which in turn causes a change in the refractive index of the material, detection of changes in the refractive index by the optical system caused by acoustic waves and converting them into an electrical signal. Or, in turn, laser pulses are used to generate acoustic waves. In these systems, a short laser pulse is focused on the surface of the material, and part of the pulse energy is converted into thermal energy, which creates an acoustic wave. This acoustic wave then propagates within the material and can be detected using an optical system that measures the changes in refractive index caused by the wave. The optical-acoustic method has a number of advantages: the optical system allows you to accurately determine the location of the defect, since it can record changes in the refractive index with high accuracy, the optical system can operate in a wide range of temperatures and humidity, which makes it very universal, this also includes the absence the need for direct contact with the object, ease of use and the ability to integrate with automated systems.

However, this system also has disadvantages. One of them is the high cost of equipment. In addition, the optical-acoustic system requires certain skills and experience for proper configuration and use [Boytsov B.V., Vasilyev S.L., Gromashev A.G., Yurgenson S.A. Non-destructive testing methods used for structures made of promising composite materials // Proceedings of MAI. 2011. No. 49.; Sokolovskaya Yu. G., Podymova N. B., Karabutov A. A. Laser optical-acoustic method for quantitative assessment of the porosity of carbon fiber reinforced plastics based on measuring their acoustic impedance // Acoustic Journal. – 2020. – T. 66. – No. 1. – pp. 86-94.]

The disadvantages of the methods include: high cost of equipment, inconsistency in ensuring high accuracy of measurements, especially when working with samples with complex geometry or high porosity, the need for highly qualified and special training of personnel, the need to calibrate equipment, which requires additional time and resources, possible limited mobility, as well as dependence of the effectiveness of methods on temperature, humidity and other external factors.

An acoustic-emission method for diagnosing products made of composite materials based on carbon fiber and a device for its implementation are known (RF patent No. 2599327 (C1), publication dated 0.10.2016). The acoustic-emission method includes installation of acoustic transducers on the product, operating in reception and emission modes, calibration, reception, registration and evaluation of acoustic emission signals, digitization of signals, their pre-processing, filtering interference, determining the time intervals between the arrival of each signal at the acoustic transducers, determination by the difference in arrival times of the coordinates of sources of acoustic emission signals. In the control zone, a piezo antenna is installed from transducers, the zone is divided into sectors, into which an acoustic transducer of the signal simulator is sequentially installed along an arc of a circle with a radius of at least half the minimum distance between the acoustic transducers, the minimum amplitude of the simulator generator is set, and the arrival times of the acoustic emission signals are determined to construct a velocity hodograph , after which a matrix of arrival time differences is constructed using the hodograph, and the location errors of the simulator signals Δux, Δuy are calculated in accordance with the expressions.

where x _loc , y _loc are the coordinates of the acoustic emission calibration signals, calculated from the matrix of arrival time differences;
x _p , y _p - real coordinates of the installation location of the simulator acoustic transducer,
Moreover, if the error exceeds the permissible value, the amplitude of the simulator generator signals is increased until the location error is within the permissible value, then, based on the recorded amplitude of the acoustic emission signals in each channel, their selection thresholds are set, after which the test object is loaded, registered at the same time the arrival times of acoustic emission signals are compared with matrix values and the coordinates of defect sources are judged from the closest of them.

The disadvantages of the method include: the requirement for the use of expensive equipment, the need for high qualifications and special training of personnel, possible limited mobility, difficulty in interpreting data, limited use in conditions of strong noise or vibration that can interfere with measurements, and the lack of a uniform determination method type of defect according to the parameters of acoustic emission signals, which in some cases can lead to the impossibility of detecting all types of defects, especially those that are small in size or have a specific shape.

There is a known method for acoustic-emission monitoring of defects in composite structures based on carbon fiber (RF patent No. 2674573 (C1), publication dated December 11, 2018). The method includes installing acoustic transducers on the sample, operating in receiving and emitting modes, calibration, static loading of the samples with a stepwise change in the load, locating sources of acoustic emission signals, which are used to judge the presence and coordinates of defects in the sample. Initially, stepwise static loading of several carbon fiber samples with the same stress concentrator is carried out until they are completely destroyed, the values of the median amplitudes of signals from the concentrator region and their structural coefficients are recorded at each loading step, and the threshold values for these parameters are calculated using the formulas

Where - respectively maximum and minimum values of structural coefficients at minimum and maximum load, determined by the formula

where m is the number of recorded signals at the i-th loading interval; D2, D3 - sets of wavelet decomposition coefficients of the 2nd and 3rd levels of detail, obtained at the sampling frequency of the original signal

SMed=[Med(a)max-Med(a)min]⋅0.1,

where Med(a)max, Med(a)min are the maximum and minimum values of the median amplitude of acoustic emission signals at maximum and minimum loads, and then static loading of the tested carbon fiber structure is carried out, the values of the median amplitudes of signals and structural coefficients are recorded, and they are compared with threshold values and with a simultaneous decrease in the structure coefficient and an increase in the median amplitude of a given group of signals, the presence of a defect and its coordinates are judged.

The disadvantages of this method are: the need to use several carbon fiber samples with the same stress raiser for their complete destruction under stepwise static loading, which requires significant time and material costs, and also creates additional difficulties in data processing and analysis.

There are known technical solutions in which there is a proven assumption that noise vibrations can carry useful information. Bounded bodies, as is known, are characterized by a certain set of natural frequencies. If there are forced harmonic oscillations in the medium, then when their frequency coincides with one of the natural frequencies of the object under study, resonance phenomena are observed, which can be used, for example, to solve problems of monitoring and diagnostics of products made of composite materials, where under the influence of noise, if in them If there are spectral components that coincide with any natural frequencies of the object, a field of standing waves can be formed in the latter, the frequency of each of which coincides with one of the natural frequencies. Despite the relatively small amplitude of these waves, a technique has been developed that allows them to be isolated from the noise field [Yu. I. Kolesnikov et al. On diagnosing the state of structural elements of structures using the noise field (according to physical modeling data) //Physical and technical problems of mineral development . – 2012. – No. 1. – P. 3-11; Fedin K.V., Kargapolov A.A., Kolesnikov Yu.I. Influence of slot-like defects on the field of standing waves formed in a fixed beam under the influence of acoustic noise // Interexpo Geo-Siberia. – 2012. – T. 1. – No. 2. – P. 88-92.; Kolesnikov Yu. I., Fedin K. V., Kargapolov A. A. On determining the natural frequencies and vibration modes of pipelines based on acoustic noise // Interexpo Geo-Siberia. – 2012. – T. 1. – No. 2. – P. 158-162.; Kolesnikov Yu. I. et al. On diagnosing the loss of stability of pipeline supports using acoustic noise //Physical and technical problems of mineral development. – 2012. – No. 4. – P. 59-67.; Fedin K. V., Klimontov V. V. A method for determining bone tissue density based on the isolation of standing waves from microseisms of the peripheral skeleton. – 2021].

The disadvantages of the method include the following: the requirement for specialized equipment and software, the need to train personnel, which may require additional training and practice, in addition, this technique may be limited in application depending on the type of object under study and noise characteristics, as well as the results , obtained using this technique may have some error or uncertainty, which may limit its use in some applications.

Disclosure of the invention.

The problem that the claimed method is aimed at solving is in the development of a non-destructive testing method based on the use of standing waves arising in a noise field, recorded by appropriate equipment and placed on the surface of a composite material product.

The technical result is a developed method that makes it possible to identify hidden defects in composite materials with high accuracy and speed of measurements, which does not require pre-treatment of the surface of the material under study and without the use of artificial seismic-acoustic noise.

A method for determining hidden defects in composite materials by acoustic noise, which consists in recording noise records with a high sampling frequency on the product under study - 65536 samples per block. The noise recording is divided into blocks. A block is a segment of noise recording that is used for analysis. Each block consists of a certain number of samples. Each block is then divided into fragments, which are used to calculate the amplitude spectrum. The amplitude spectra are calculated, then for each observation point the amplitude spectra of all fragments are averaged, the presence of standing waves is established, and a change in compression-expansion type modes indicates the presence of defects in the product under study.

Noise recordings are divided into fragments of 65,536 counts each, and the amplitude spectra of these fragments are calculated and accumulated. Spectral stacking is a process used in signal processing to improve the frequency resolution and amplitude accuracy of spectral components. The accumulation of spectra is based on dividing the original signal into several fragments (blocks) and performing the Fourier transform on each of them. The results of these transformations are then combined to provide more accurate information about the spectrum of the signal. The vertical component of seismic noise of compression-expansion modes is also recorded by detailed profiling in a localized area, and the location of defects is clarified.

The amplitude spectra are calculated using special signal processing algorithms (Fourier transform). Averaging of amplitude spectra is carried out in order to obtain a more accurate picture of the distribution of amplitudes in noise. If there are defects in the product under study, this can manifest itself in the form of a change in compression-expansion type modes, which is what this method allows us to detect. The Fourier transform is used to move from a time representation of a signal to a frequency representation.

The conversion formula is as follows:

where F(ω) is the frequency spectrum of the signal, f(t) is the time signal, ω is the frequency.

Additionally, the power of the recorded signal can be calculated by integral summation of its spectrum over a fixed time interval. A fixed time interval is an arbitrary predetermined period of time, which is selected depending on the task (1 second, etc.). The formula for calculating signal power through the integral summation of the amplitude spectrum is as follows:

P= df,

where P is the signal power, S(f) is the spectral density of the signal, f ₁ and f ₂ are the initial and final frequencies of the spectrum summation range.

The claimed method does not require pre-treatment of the surface of the material under study and the use of any artificial sources (oscillations, radiation, etc.), which makes the proposed method universal in relation to the objects under study, highly productive and safe for the operator.

Fig. 1. The object under study (prosthetic foot part) with a piezoceramic sensor.

Fig. 2. “Grid” of measurements of the moving sensor.

Fig.3. Comparison of amplitude-frequency spectra for a whole sample and a sample with a crack

Fig. 4. The amplitude-frequency spectra are shown for the whole sample and a sample with a crack (single specimen).

Implementation of the invention.

A batch of prosthetic leg parts made from composite materials was used as research objects. In the described series of experiments, recording of seismoacoustic noise was carried out in the laboratory.

The vertical components of seismoacoustic noise at the measurement point are recorded. Columns were used as a noise source, to which a “white noise” signal type was supplied up to 20 kHz, which speeds up the data collection process; however, the use of columns is not mandatory for the technique. The amplitude-frequency response of the speakers according to the passport works up to 20 kHz.

Then, the noise field is recorded on a mobile receiver, which is based on a TsTS-19 piezoceramic disk with a diameter of 2 mm and a thickness of 1 mm (Fig. 1). The receiver is moved over the object under study made of composite material with the following “grid” of measurements: 2 longitudinal tracks with 10 points were passed along in increments of 2 cm and 2 transverse tracks with 3 points were passed in increments of 1 cm (Fig. 2). Signals from the sensor are transmitted to one channel of a two-channel digital oscilloscope B-421, then digitally recorded on the hard drive of a personal computer for subsequent processing.

The essence of the method is the accumulation of a large number of amplitude spectra of noise recordings, as a result of which sequences of peaks corresponding to families of standing waves of different types appear on the averaged (or accumulated) spectra.

The standing wave diagnostic method was as follows:

1. Registration of noise records on the object under study to identify standing waves in them using a piezoceramic sensor moved along the measurement “grid”. A measurement “grid” is a coordinate system that is used to describe in what manner point-by-point measurements were taken from the object under study. It is a set of points on the object under study, where each point has its own coordinates (for example, x and y), in our case these are 2 longitudinal tracks with 10 points along with a step of 2 cm and 2 transverse tracks with 3 points in increments of 1 cm;

2. Dividing noise records into blocks of 65536 samples (the maximum number of samples in the B-421 oscilloscope). The noise recording is divided into many fragments, each of which contains 65536 samples. “Sampling” means the process of measuring or recording data, in this context it means that for each piece of noise, its amplitude and phase have been measured;

3. Fourier transform for each block and averaging of the resulting amplitude-frequency spectra by summing them;

4. Construction and comparison of the total spectra of defective and intact samples in order to identify a shift in the amplitude-frequency spectrum and the formation of new resonance peaks as an assessment of the presence of defects.

The correspondence of the identified regular peaks to standing waves of vertical compression-tension of products made of composite material, and not to standing waves of other types (for example, shear), is due to the use of sensors that record predominantly the vertical component of acoustic noise at the measurement point (Figure 3).

By comparing the total spectra of a whole sample and a sample with a crack in a single specimen, one can observe the appearance of additional peaks corresponding to the resonant frequencies of the defective product, which indicates the appearance of additional reflection boundaries due to the occurrence of breaks in its structure. Using the example of comparing the total spectra of a whole sample and a sample with a crack, we can conclude that a downward shift in the resonance frequencies of one of them may be due to the fact that this sample has a lower quality factor compared to the other, which indicates that that the object may contain defects that can cause energy loss and impair the ability of the product under study to retain it, which, in turn, can lead to reduced strength of the product (Figure 4).

The experimental results obtained showed that the standing wave method can be successfully used to diagnose hidden defects in composite materials. Standing waves can be separated from a noise field by accumulating a large number of amplitude spectra of noise signals. Maps of amplitude-frequency distributions were obtained.

The final result of the examination is expected to provide the amplitude-frequency distribution of the lowest modes for the entire product. The time required to conduct field work and interpret data depends on the detail of the measurements. Presumably, examining one part using this technique to obtain the final result will take no more than 1 hour.

Thus, the fundamental possibility of using the technique of separating elastic standing waves from seismoacoustic noise on the studied objects made of composite material was demonstrated.

Claims (6)

1. A method for identifying hidden defects in composite materials by acoustic noise, which consists in recording noise records with a sampling frequency of 20 Hz on the test product made of composite material, the noise record is divided into blocks, which consist of a certain number of samples, then each block is divided into fragments with a duration of 65536 counts and calculate and accumulate the amplitude spectra of these fragments according to the formula where F(ω) is the frequency spectrum of the signal, f(t) is the time signal, ω is the frequency, then, for each observation point, the amplitude spectra of all fragments are averaged, the presence of standing waves is established, and a change in compression-expansion type modes indicates the presence of defects in the composite material. 2. The method according to claim 1, characterized in that the vertical component of the seismic noise of the compression-expansion modes is pre-registered by detailed profiling in a localized area, and the location of the defects is specified. 3. The method according to claim 1, characterized in that it can be supplemented by calculating the power of the recorded signal by integral summation of its spectrum over a fixed time interval.