RU2537747C1 - Acoustic-emission method to diagnose metal structures - Google Patents

Abstract

FIELD: testing equipment.

SUBSTANCE: use: for diagnostics and non-destructive check of metal structures. The substance of the invention consists in the fact that parameters of acoustic emission signals are received, recorded and assessed at the moment of structure loading, acoustic signals are digitised, they are preliminarily processed, noise is filtered, at the same time they first establish critical values of the load P_cr and regression coefficient k_rc, which characterises variation of the number of acoustic emission signals to variation of the load for a non-defect structure, then the structure is loaded to the value of the load exceeding the working load by (5…10)%, at the same time they record the number of signals and the load of the linear section of stationary acoustic emission, at the same time they record the regression coefficient k₀, afterwards the structure is loaded by cyclic load, the amplitude value of which is exceeded gradually by (2…5)%, and whenever exceeding reaches (15…20)% above the working load, loading is stopped, if in the process of control k₀<k_rc, then the structure is considered to be free of defects, and if the value k₀>k_rc the structure is rejected.

EFFECT: improved validity of acoustic-emission control of metal structures.

3 cl, 4 dwg, 4 tbl

Description

The invention relates to the field of technical diagnostics and non-destructive testing of metal structures of a wide profile, including welded structures, pressure vessels, pipelines, aircraft and railway structures, bridges, etc. using the acoustic emission method.

A known method for diagnosing metal bridge structures and a device for its implementation (RF patent No. 2240551, IPC ⁷ G01N 29/04. Method for diagnosing bridge metal structures and a device for its implementation / Stepanova L.N., Muraviev V.V. and others. priority of June 20, 2001, Bull.№32, 2004, adopted as an analogue), which includes receiving, recording and evaluating parameters of acoustic emission signals at the time of loading a bridge metal structure, for example, by a passing train, digitizing acoustic signals, calculating the spec ra acoustic signals, their pretreatment, noise filtering, logging the arrival time of the acoustic signals and calculating coordinates thereon developing defects. In addition, simultaneously with the registration of signals from acoustic transducers, dynamic deformation is recorded, and the main parameters of acoustic signals, coordinates of developing defects and their spectral characteristics are recorded at the time of maximum mechanical deformation of the structure.

The disadvantage of this method is that the recording of informative parameters of acoustic emission signals in the analogue is carried out only at the time when the design reaches the maximum strain. However, since it is impossible to determine the deformation value in advance, it is impossible to determine the maximum of deformations, and, therefore, set a threshold on them and find the time period during which it is necessary to record acoustic emission signals. Therefore, the response threshold for deformations is set according to the value of the dynamic deformation measured before the moment of receiving the acoustic signal. As a result of this, acoustic emission information may be lost, as a result of which the localization of defects is determined with a large error. In addition, the process of finding the maximum of deformations and determining the period of time during which the recording of acoustic emission information is very complicated and takes a long period of time. This leads to a decrease in the speed of the device, to the loss of information and a decrease in the reliability of the measurement results.

Closest to the proposed solution is a method for diagnosing metal structures and a device for its implementation (Pat. RF No. 2339938, IPC ⁷ G01N 29/04, priority dated 02.14.2007, Bull. No. 33, 2008, adopted as a prototype), including receiving, recording and evaluating parameters of acoustic emission signals at the time of loading a metal structure with a cyclic load, recording dynamic deformations, digitizing acoustic signals, their preliminary processing, filtering interference, recording the time of arrival of acoustic signals calculating coordinates thereon developing defects. Furthermore, simultaneously with the registration of the dynamic deformation is performed registration time of arrival of the acoustic emission signal with a minimum U _pormin threshold equal to the minimum noise level and the maximum threshold is equal to U _pormax = (U _pormin + U _ext), where U _ext - add value threshold equal to the difference between the minimum and maximum noise levels, and the measured deformation is used to determine the phase ξ of the load by the formula:

$$\xi =arcsin\frac{\epsilon }{{\epsilon }_{max}}$$

where ε is the current strain; ε _max - maximum deformation,

and according to the load phase, clustering is performed for each acoustic signal, and the target clustering function is found from the condition:

; $${\alpha }_{\xi }=C_{\xi }\cdot \varphi \left(\lambda \right)$$

$$\alpha =\sqrt{{\alpha }_r{}^2}+{\alpha }_{\xi }{}^2$$

; $${\alpha }_{\xi }=C_{\xi }\cdot \varphi \left(\lambda \right)$$

- геометрический признак, равный отношению площади области локализации сигнала акустической эмиссии, полученной при минимальном и максимальном значениях порогового уровня, и площади перекрытия области локализации сигнала и кластера; C_ξ - «весовой» коэффициент фазы нагрузки; C_r - «весовой» коэффициент геометрического признака; S_rлло - площадь области локализации сигнала; S_Ki∩rлло - площадь перекрытия области локализации сигнала в кластере, если число сигналов в таком кластере превышает заданный порог по количеству акустических сигналов, то это соответствует наличию дефекта с координатами, равными координатам центра кластера.where φ (λ) is the normal distribution function; λ is the normalized deviation of the load phase; $${\alpha }_r=\frac{C_r\cdot S_{r\text{A.}ll\mathrm{about}}}{S_{Ki}{\cap }_{r\text{A.}l\mathrm{about}\mathrm{to}}}$$

- a geometric feature equal to the ratio of the area of ​​the localization region of the acoustic emission signal obtained at the minimum and maximum values ​​of the threshold level and the overlap area of ​​the region of localization of the signal and cluster; C _ξ is the “weight” coefficient of the load phase; C _r is the “weight” coefficient of the geometric feature; S _rlo - area of ​​the signal localization region; S _Ki∩rlo - the overlap area of ​​the signal localization region in the cluster, if the number of signals in such a cluster exceeds a predetermined threshold by the number of acoustic signals, then this corresponds to the presence of a defect with coordinates equal to the coordinates of the center of the cluster.

The disadvantage of the method adopted for the prototype is the inability to control and change (if necessary) the loading speed. In addition, such tests are always carried out under cyclic loading, the period and amplitude of which are selected separately and are not related to the parameters of acoustic emission, that is, the destruction process and the safety of the tests are not controlled in any way by the recorded characteristics of acoustic emission signals directly related to the process of destruction of the diagnostic object.

When developing the inventive acoustic emission method for diagnosing metal structures, the task was to increase the reliability of the monitoring results due to the fact that the parameters of the recorded acoustic emission signals establish the optimal loading parameters for detecting defects: type of loading, rate of change of load, frequency of cyclic loading, amplitude value of load . In addition, in the claimed method (in contrast to the prototype), the reliability of detecting developed defects is increased due to the use of accumulated acoustic emission signals during repeated loading with varying amplitude and frequency of the load.

The problem is solved due to the fact that in the proposed acoustic emission method for diagnosing metal structures, which consists in the fact that they receive, register and evaluate the parameters of acoustic emission signals at the time of loading of the structure, digitize the acoustic signals, their preliminary processing, filtering interference. In addition, first establish the critical values of the load P _cr and the regression coefficient k _cr characterizing the change in the number of acoustic emission signals to the change in load for a defect-free design, then the design is loaded to a load value (5 ... 10)% higher than the working one, and the number is recorded signals and the load of the linear section of stationary acoustic emission, the regression coefficient k _{0 is} recorded, after which the structure is loaded with a cyclic load, the amplitude value of which is increased gradually by (2 ... 5)%, and when the excess by (15 ... 20)% of the workload is reached, loading is stopped, if during the control process k ₀ <k _cr , then the structure is considered defect-free, and if k ₀ > k _{cr, the} structure is rejected .

, где ΔP_Б, ΔN_Б - соответственно изменения нагрузки и числа сигналов акустической эмиссии бездефектной конструкции.In addition, the critical regression coefficient is determined by the formula $$k_{\mathrm{to}R}=\left(3...5\right)\cdot \frac{\Delta N_B}{\Delta P_B}$$

, where ΔP _B , ΔN _B are, respectively, changes in the load and the number of acoustic emission signals of a defect-free design.

In addition, the loading speed and frequency are regulated depending on the recorded activity and attenuation of acoustic emission signals so that the number of unregistered signals does not exceed 1-5%.

Figure 1 shows a graph of the dependence of the applied load on time in the process of acoustic emission control.

Figure 2 shows a graph of the dependence of the number of acoustic emission signals on the load applied to the structure under static loading.

In FIG. Figure 3 shows the time dependences of the load and activity of acoustic emission signals.

Figure 4 shows the localization of acoustic emission signals when testing the side frame of the truck wagon.

The proposed method is as follows. First set critical load value P _cr equal to the design maximum working load P _slave, and the critical regression coefficient k _cr equal to:

$$k_{\mathrm{to}R}=\left(3...5\right)\cdot \frac{\Delta N_B}{\Delta P_B}$$

where ΔP _B , ΔN _B are, respectively, changes in the load and the number of acoustic emission signals during testing of defect-free structures that were not in operation.

. В процессе статического нагружения регистрируют число сигналов N и нагрузку P₀ начала линейного участка стационарной акустической эмиссии по графику изменения N(P) и k₀=tg(α) (фиг.2):Then, static loading is carried out (Fig. 1), applying a monotonically increasing load to the structure to the value of P _article , which is (5 ... 10)% higher than the maximum working load. At the same time, receiving, recording and evaluating the parameters of acoustic emission signals, digitizing acoustic signals, their preliminary processing, filtering interference. The number of acoustic emission signals and the load of the linear portion of the stationary acoustic emission are recorded (FIG. 2) and the regression coefficient is determined $$k_0=\frac{\Delta N}{\Delta P}$$

. In the process of static loading, the number of signals N and the load P _{0 of the} beginning of the linear section of stationary acoustic emission are recorded according to the graph of changes in N (P) and k ₀ = tg (α) (Fig. 2):

, $$N=k_0\left(P-P_0\right)$$

,

сигналов акустической эмиссии (фиг.3):If the load at the beginning of the linear section P _{0 is} greater than the critical P ₀ > P _cr or the coefficient k ₀ <k _cr , then carry out a cyclic loading (figure 1). The frequency of cyclic loading is regulated depending on the maximum activity $$\dot{N}$$

acoustic emission signals (figure 3):

, $$V_{l+\mathrm{one}}=v_0\cdot \frac{{\dot{N}}_{\mathrm{to}R}}{{\dot{N}}_j+\frac{{\dot{N}}_{\mathrm{to}R}}{2}}$$

,

where l is the sequence number of the cycle; V ₀ - the initial frequency corresponding

. $$\frac{\mathrm{one}}{2\cdot \Delta t}=(\mathrm{30..40})c$$

.

и их затухания следующим образом (фиг.3). Устанавливают начальную скорость изменения нагрузки, равной $$V_0=\frac{P_{ст}}{\Delta t}$$

. На интервале времени t_j определяют активность сигналов акустической эмиссии $${\dot{N}}_j$$

. При этом скорость изменения нагрузки V_j+1 в следующий момент времени t_j+1 устанавливают равной:During cyclic loading, the rate of change of load V is regulated depending on the activity of acoustic emission signals $$\dot{N}$$

and their attenuation as follows (figure 3). Set the initial rate of change of load equal to $$V_0=\frac{P_{\mathrm{from}t}}{\Delta t}$$

. On the time interval t _j determine the activity of acoustic emission signals $${\dot{N}}_j$$

. The rate of change of load V _{j + 1} at the next time t _{j + 1 is} set equal to:

$$V_{j+\mathrm{one}}=v_0\cdot \frac{{\dot{N}}_{\mathrm{to}R}}{{\dot{N}}_j+\frac{{\dot{N}}_{\mathrm{to}R}}{2}}$$

- предельная активность сигналов акустической эмиссии, которая может быть зарегистрирована в данной конструкции и зависящая от коэффициента затухания δ:Where $${\dot{N}}_{\mathrm{to}R}$$

- the ultimate activity of acoustic emission signals, which can be recorded in this design and depending on the attenuation coefficient δ:

, $$N_{\mathrm{to}R}=k_3\cdot \delta$$

,

where k ₃ is the attenuation coefficient equal to the ratio of the maximum amplitude of the acoustic emission signals to the detection threshold. This loading rate allows you to record the maximum number of acoustic emission signals in the control object and to obtain more reliable information about its state.

сигналов акустической эмиссии, поэтому для интервала времени t_j устанавливают низкую скорость изменения нагрузки V, а в интервал времени t_j наблюдается низкая активность сигналов акустической эмиссии, поэтому устанавливают высокую скорость изменения нагрузки для интервала времени t_j+1.Figure 3 in the time interval t _j-1 there is a high activity $$\dot{N}$$

acoustic emission signals, therefore, for the time interval t _j set a low rate of change of load V, and in the time interval t _j there is a low activity of acoustic emission signals, therefore, set a high rate of change of load for the time interval t _{j + 1} .

If the load at the beginning of the linear section P _{0 is} less than the critical P _{0 <} P _cr and the coefficient k ₀ > k _cr , then loading is stopped and the structure is rejected.

до нагрузки, превышающей максимальную рабочую нагрузку на (15…20) %.The maximum load of the cycle is gradually increased by (2 ... 5)% every Δn cycles, depending on the coefficient $$k_0=\frac{\Delta N}{\Delta P}$$

up to a load exceeding the maximum working load by (15 ... 20)%.

In the process of cyclic loading, the number of signals N and the load P, which determine the load P _{0 of the} beginning of the linear section of stationary acoustic emission and the coefficient k _{0 of the} ratio of the change in the number of acoustic emission signals ΔN to the load change ΔP, are recorded and accumulated in the time interval corresponding to Δn cycles. If after Δn cycles the coefficient k ₀ <k _cr , then the amplitude of the cyclic load is increased by (2 ... 5)%. If when the maximum load on (15 ... 20)% above the maximum workload and the inequality k ₀ <k _cr, the structure is considered a defect-free.

, и с учетом динамики движения принята равной $$P_{раб}=P_{ось}+0,6\cdot P_{ось}=368кН$$

.The proposed method was implemented in the control of the side frames of the box-shaped section of the freight car truck. The maximum working load of the structure with the permitted static axle load equal to $$P_{\mathrm{about}\mathrm{from}b}=230\mathrm{to}N$$

, and taking into account the dynamics of movement is taken equal $$P_{R\mathrm{but}b}=P_{\mathrm{about}\mathrm{from}b}+0,6\cdot P_{\mathrm{about}\mathrm{from}b}=368\mathrm{to}N$$

.

, а критический коэффициент, определенный экспериментально в предварительных испытаниях новых боковых рам без усталостных дефектов, равным:The critical load value was set equal to $$P_{\mathrm{to}R}=P_{R\mathrm{but}b}=368\mathrm{to}N$$

, and the critical coefficient determined experimentally in preliminary tests of new side frames without fatigue defects is equal to:

$$k_{\mathrm{to}R}=0,43\frac{\mathrm{one}}{\mathrm{to}H}.(\mathrm{one})$$

и одновременно проводили прием, регистрацию и оценку параметров сигналов акустической эмиссии, оцифровку акустических сигналов, их предварительную обработку, фильтрацию помех, регистрацию времени прихода акустических сигналов и вычисление по нему координат развивающихся дефектов (Диагностика объектов транспорта методом акустической эмиссии /Серьезнов А.Н., Степанова Л.Н. и др. - / Под ред. Л.Н. Степановой, В.В. Муравьева - М.: Машиностроение, 2004, с. 74-81). Локализация сигналов акустической эмиссии в исследуемой боковой раме коробчатого сечения тележки грузового вагона приведена на фиг.4.Then we conducted a static loading of the side frame with a monotonously increasing load to a value $$P_{\mathrm{from}t}=387\mathrm{to}N$$

and simultaneously received, recorded and evaluated the parameters of acoustic emission signals, digitized acoustic signals, their preliminary processing, filtering interference, recording the time of arrival of acoustic signals and calculating the coordinates of developing defects from it (Diagnostics of transport objects by acoustic emission / A. Seryozhnov, Stepanova L.N. et al. - / Under the editorship of L.N. Stepanova, V.V. Muravyev - M .: Mechanical Engineering, 2004, p. 74-81). The localization of acoustic emission signals in the studied lateral frame of the box section of a freight car truck is shown in Fig. 4.

. В процессе нагружения скорость изменения нагрузки V регулировали в зависимости от активности сигналов акустической эмиссии $$\dot{N}$$

(таблица 1).In the process of preliminary tests, the attenuation coefficient δ = 5851 1 / s and a coefficient equal to the ratio of the maximum amplitude of the acoustic emission signals to the detection threshold k ₃ = 46 were established. The maximum activity of acoustic emission signals was $${\dot{N}}_{\mathrm{to}R}=127\mathrm{one}/c$$

. During loading, the rate of change in load V was regulated depending on the activity of acoustic emission signals $$\dot{N}$$

(Table 1).

.In the process of static loading determine the load of the beginning of the linear section of stationary acoustic emission, equal to P ₀ = 362 kN. In this case, the coefficient corresponding to the ratio of the change in the number of acoustic emission signals caused by the increment of the load is $$k_0=0.18\frac{\mathrm{one}}{\mathrm{to}H}$$

.

Table 1 Time, activity and rate of change of load during static tests

п/п $$\dot{N}$$

p / p 0 one 2 3 four 5 6 7 8 9 10 eleven 12 13 .. t _i , c 0 0.5 1,0 1,5 2.0 2,5 3.0 3,5 4.0 4,5 5,0 5.5 6.0 6.5 .. $${\dot{N}}_i$$

,
1 / c 0 0 10 eighteen 17 16 19 21 fourteen 13 eighteen 10 eighteen 13 .. V _i
kH / c 12.9 12.9 12.9 11.0 9.8 9.9 10.0 9.7 9,4 10,4 10.5 9.8 11.0 9.8 ..

From the formula (1) it follows that the inequality holds and the coefficient k ₀ <k _cr. Then, cyclic loading of the side frame was carried out. The frequency of cyclic loading was regulated depending on the maximum activity of acoustic emission signals:

, $$V_{l+\mathrm{one}}=v_0\cdot \frac{{\dot{N}}_{\mathrm{to}R}}{{\dot{N}}_j+\frac{{\dot{N}}_{\mathrm{to}R}}{2}}$$

,

where l is the sequence number of the cycle.

Table 2 shows data on the frequency, activity and the number of loading cycles of the side frame.

table 2 Dependence of frequency and activity on the number of loading cycles Cycle number one 2 3 four 5 6 7 8 9 10 eleven ..

,
1/c $${\dot{N}}_i$$

,
1 / c 5 8 12 fifteen four 124 25 21 29th 5 42 .. v ₁ , 1 / c 0.20 0.37 0.36 0.34 0.32 0.38 0.14 0.29 0.30 0.27 0.37 ..

The maximum load of the cycle was increased every Δn = 10 cycles by 18 kN to a maximum load of 441 kN. At the same time, the number of signals N and the load P were determined and accumulated on the basis of Δn = 10 cycles and the load P was used to determine the load P _{0 of the} beginning of the linear section of stationary acoustic emission and the coefficient k _{0 of the} ratio of the change in the number of acoustic emission signals to the load change (table 3).

Since after 30 cycles the coefficient k ₀ = 0.48 1 / kN exceeds the critical value equal to k _cr = 0.43 ¹ / _kN , and the load P ₀ = 321 kN below the critical value P _cr = 368 kN, the side frame is rejected.

Table 3 The dependence of the load P _{0 the} beginning of the linear section of stationary acoustic emission and the coefficient K ₀ changes in the number of signals to change the load on the number of the loading cycle Cycle number 10 twenty thirty P ₀ , kN 351 346 321 k ₀ , 1 / kN 0.32 0.39 0.48

The method was tested experimentally on 12 defective and defect-free side frames of the box-shaped section of the trucks of freight cars. The measurement results of the diagnostic parameters of the twelve side frames are presented in table 4.

The parameters of the recorded acoustic emission signals establish the optimal loading parameters for detecting defects. The rate of change of the load and the frequency of cyclic loading are controlled depending on the activity of the acoustic emission signals. With an increase in activity, the rate of change in load is reduced, and with a decrease in activity, it is increased. Thus, unlike the prototype, the probability of missing acoustic emission signals from developing defects due to the superposition of acoustic emission signals, both under static and cyclic loading, is reduced. In addition, the inventive method allows you to detect defects that when loading monotonically increasing load was not detected. The increase in the probability of detecting a defect, in comparison with the prototype, is achieved due to the accumulation of acoustic emission signals during cyclic loading and a gradual increase in the amplitude value of the load. At the same time, optimal time is ensured, which is especially important when organizing flow control of structures in a depot.

Table 4 Results diagnosing side frames freight cars at the value of the basic static load of 387 kN and k _cr = 0.43 Head Part No. Number of signals k ₀ cm, 1 / kN k _0c , 1 / kN kN The number of cycles Control result T103 6 0.015 0,07 441 40 Fit 8752 10 0,026 0.16 441 40 Fit 19644 four 0.08 0.11 441 40 Fit 25036 8 0.17 0.27 441 40 Fit 53104 9 0.18 0 441 40 Fit 25969 16 0.4 0.66 405 twenty Marriage T91013 7 0.18 0.48 441 40 Marriage 8172 17 0.68 - - - Marriage 79151 16 0.32 0.55 387 10 Marriage 6492 19 0.44 - - - Marriage 6667 17 0.36 1,62 405 twenty Marriage 66991 twenty 0.44 - - - Marriage

Claims (3)

1. The acoustic emission method for diagnosing metal structures, including receiving, recording and evaluating the parameters of acoustic emission signals at the time of loading the structure, digitizing acoustic signals, their preliminary processing, filtering interference, characterized in that the critical load values P _cr and the regression coefficient are first set k _cr characterizing a change in the number of acoustic emission signals to a change in load for a defect-free design, then the design is loaded to a load value loads exceeding the working one by (5 ... 10)%, in this case, the number of signals and the load of the linear section of the stationary acoustic emission are recorded, the regression coefficient k _{0 is} recorded, after which the structure is loaded with a cyclic load, the amplitude value of which is gradually increased by (2 ... 5 )%, and when the excess by (15 ... 20)% of the workload is reached, loading is stopped, if during the control process k ₀ <k _cr , then the structure is considered defect-free, and if k ₀ > k _{cr the} structure is rejected. 2. The method according to claim 1, characterized in that the critical regression coefficient is determined by the formula $$k_{\mathrm{to}R}=\left(3\mathrm{..}5\right)\cdot \frac{\Delta N_B}{\Delta P_B}$$

, where ΔP _B , ΔN _B are, respectively, changes in the load and the number of acoustic emission signals of a defect-free design. 3. The method according to claim 1, characterized in that the speed and frequency of loading is regulated depending on the recorded activity and attenuation of the acoustic emission signals so that the number of unregistered signals does not exceed 1-5%.

Images