CN108061759A - A kind of Reason of Hydraulic Structural Damage recognition methods based on piezoelectric ceramics - Google Patents

Abstract

The invention discloses a damage identification method for hydraulic concrete structures based on piezoelectric ceramics, which includes an active monitoring process and a passive monitoring process. The present invention utilizes the electric signal monitored by the piezoelectric ceramic sensor, and through a series of analysis and processing, it can effectively identify the damage of hydraulic concrete, especially the crack damage, that is, it can accurately determine whether there is damage or not, and the occurrence of damage location and extent of damage. The invention can use the piezoelectric ceramic monitoring technology in the dam safety monitoring system to identify the damage of hydraulic concrete, has small volume, high sensitivity and fast response speed, and can be widely used in the safety monitoring system.

Description

A damage identification method for hydraulic concrete structures based on piezoelectric ceramics

technical field

The invention relates to a damage identification method for hydraulic concrete structures, in particular to a damage identification method for hydraulic concrete structures based on piezoelectric ceramics.

Background technique

Due to self-weight load, hydrostatic pressure, wind load, temperature change, and the internal force of the surrounding rock mass, the hydraulic concrete structure must also bear extreme loads, such as earthquakes, severe floods, droughts and other unfavorable loads. Influenced by defects in construction, construction and management, damage will inevitably occur. In order to ensure the healthy operation of hydraulic concrete structures, it is very important to identify the damage of its state. Piezoelectric ceramics are widely used in structural damage detection and identification due to their low cost and high sensitivity.

For concrete dams, the main form of damage is cracks. Early cracks in concrete are difficult to find. If tiny cracks are ignored and allowed to develop, they may gradually expand under the influence of water pressure, physical and chemical erosion and the environment. form macroscopic cracks. If there are macro cracks in the concrete dam, the phenomenon of stress concentration will easily occur, and then the structure of this part will be damaged, and the whole structure may eventually be destroyed. Therefore, we need to give feedback on the health status of the hydraulic concrete structure in a timely manner and put forward rationalization opinions when there is no large area of damage or the damage is very small, so as to prevent problems before they happen.

Choosing an appropriate monitoring method to monitor the health of hydraulic concrete structures such as dams and identifying structural damage as early as possible is of great significance to ensure the normal operation of the structure, reduce the loss of economic and social benefits, and protect the safety of people's lives and property.

Since the discovery of the piezoelectric effect, sensors made of piezoelectric ceramics have been widely used in various fields for damage monitoring of structures. Piezoelectric ceramics integrate sensing and driving functions, small size, sensitive response, wide frequency response range, low price, and easy tailoring. Using them as sensors for structural damage monitoring can not only identify local damage to the structure, but also affect the overall structure. Damage is identified. At the same time, it can also carry out long-term, continuous and real-time monitoring of the structure. Therefore, it is of great significance to use piezoelectric ceramic sensing technology to identify the damage of hydraulic concrete structures.

However, the signals collected in the prior art inevitably contain noise, which makes the signal-to-noise ratio of the response data small, and useful signals containing characteristic parameters are often submerged in the noise, so certain signal noise reduction methods and signal analysis methods need to be adopted. Useful information about the security state of the reaction structure is extracted. In addition, due to the large size and high degree of freedom of hydraulic structures, there will be problems of false modes in the process of modal parameter identification. Therefore, the above problems need to be improved, so that the damage identification method has practical application value.

Contents of the invention

Purpose of the invention: The purpose of the present invention is to provide a damage identification method for hydraulic concrete structures based on piezoelectric ceramics that can solve the defects in the prior art.

Technical scheme: in order to achieve this goal, the present invention adopts following technical scheme:

The damage identification method for hydraulic concrete structures based on piezoelectric ceramics of the present invention includes an active monitoring process and a passive monitoring process, which are respectively:

The active monitoring process includes the following steps:

S11: Using the moving average method to eliminate the trend item of the measured signal y(t) of the piezoelectric ceramic sensor; wherein, t is a time signal.

S12: Using the improved empirical mode decomposition method to denoise the signal, and obtain a preprocessed signal near the baseline and without noise superimposition;

S13: Decomposing the signal obtained in step S12 by using wavelet packets;

S14: Calculate the wavelet packet energy spectrum of the signal;

S15: Establish damage indicators to identify the degree of structural damage and locate the damage;

The passive monitoring process includes the following steps:

S21: Using the moving average method to eliminate the trend item of the measured signal of the piezoelectric ceramic sensor;

S22: Using the improved empirical mode decomposition method to perform signal noise reduction, and obtain a preprocessed signal near the baseline and without noise superimposition;

S23: extract the free vibration response from the signal obtained in step S22;

S24: performing modal parameter identification on the free vibration data;

S25: Establishing damage indication indicators, judging whether damage occurs and determining the degree of damage.

Further, the step S12 specifically includes the following steps:

S12.1: Obtain the lth IMF signal c _l (t), l=1,2,3,...,;

S12.2: Subtract c _l (t) from f(t) to get r _l (t), that is, r _l (t) = f(t)-c _l (t); take r _l (t) as the original The data repeats the above steps to obtain the second component c ₂ (t) of f(t) that meets the IMF condition, and so on for n times, and obtains n components of the signal f(t) that meet the IMF condition; where, f(t ) is the deterministic component of the measured signal y(t) in step S11.

Further, the step S12.1 specifically includes the following steps:

S12.11: let k=2;

S12.12: Obtain the upper envelope u _k-1 (t) and the lower envelope v _k-1 (t) by interpolating between all local maxima through the Gaussian process;

S12.13: Calculate the mean value m _k-1 (t) of the upper and lower envelopes by formula (1):

S12.14: Subtract m _k-1 (t) from f(t) to obtain a new data sequence h _k-1 (t) with low frequency removed, namely:

h _k-1 (t)=f(t)-m _k-1 (t) (2)

Wherein, f(t) is the deterministic component of measured signal y(t) in step S11;

S12.15: Determine whether k is equal to K, K is the maximum number of iterations: if yes, end; otherwise, proceed to step S12.16;

S12.16: If h _k-1 (t) meets the IMF condition, then h _k-1 (t) is the k-1th component of f(t); otherwise, set k=k+1, and then return to the execution step S12.12.

Further, the step S14 specifically includes the following steps:

S14.1: Extract the wavelet packet decomposition coefficient of the response signal through formula (3)

Among them, R(t) represents the new data series after denoising, ψ _j,h,i (t) is the wavelet packet with scale index j, position index h and frequency index i;

S14.2: Reconstruct the wavelet packet decomposition coefficients, extract signals in each frequency band, and obtain the wavelet packet energy spectrum E of the signal:

in, Represents the energy of the i-th frequency band, as shown in formula (5);

in, express the reconstructed signal,

Further, in the step S15, the process of identifying the degree of structural damage is as follows: select a signal transmitter to transmit a waveform signal, and identify the degree of structural damage by receiving the amplitude of the signal piezoelectric ceramic sensor: if the received signal If the amplitude is smaller than the amplitude under the non-destructive condition, it is determined that there is damage in the monitored area; if the amplitude of the received signal is equal to the amplitude under the non-destructive condition, it is determined that there is no damage in the monitored area The presence.

Further, in the step S23, the free vibration response is extracted from the preprocessed data by a random subtraction method.

Further, the damage indication index in the step S25 is a natural frequency index, and the jth order natural frequency index f _nj is shown in formula (6):

Among them, f _uj is the j-th order natural frequency of the non-destructive structure, and f _dj is the j-th order natural frequency of the structural damage state.

Beneficial effects: the present invention discloses a method for identifying damage to hydraulic concrete structures based on piezoelectric ceramics. Using electrical signals monitored by piezoelectric ceramic sensors, after a series of analysis and processing, the damage of hydraulic concrete can be effectively identified. Damage, especially crack damage, can accurately determine whether there is damage, as well as the location of damage and the degree of damage. The invention can use the piezoelectric ceramic monitoring technology in the dam safety monitoring system to identify the damage of hydraulic concrete, has small volume, high sensitivity and fast response speed, and can be widely used in the safety monitoring system.

Description of drawings

Fig. 1 is the flowchart of the method in the specific embodiment of the present invention;

2 is a schematic diagram of the buried position of the piezoelectric ceramic sensor during active monitoring in a specific embodiment of the present invention;

Fig. 3 is the schematic diagram of reinforced concrete beam active monitoring test system in the specific embodiment of the present invention;

Fig. 4 is the schematic diagram of the active monitoring test kerf position of reinforced concrete beam in the specific embodiment of the present invention;

Fig. 5 is the ERPS-2 received signal diagram of different damage degrees in the specific embodiment of the present invention;

Fig. 6 is a total energy index diagram of six kinds of damage working conditions in the specific embodiment of the present invention;

Fig. 7 is the natural frequency index under various damage conditions during passive monitoring in the specific embodiment of the present invention.

Detailed ways

The technical solution of the present invention will be further introduced below in combination with specific implementation methods and accompanying drawings.

This specific embodiment discloses a damage identification method for hydraulic concrete structures based on piezoelectric ceramics, as shown in Figure 1, including an active monitoring process and a passive monitoring process, respectively:

The active monitoring process includes the following steps:

S11: Using the moving average method to eliminate the trend item of the measured signal y(t) of the piezoelectric ceramic sensor; wherein, t is a time signal.

S12: Using the improved empirical mode decomposition method to denoise the signal, and obtain a preprocessed signal near the baseline and without noise superimposition;

S13: Decomposing the signal obtained in step S12 by using wavelet packets;

S14: Calculate the wavelet packet energy spectrum of the signal;

S15: Establish damage indicators to identify the degree of structural damage and locate the damage;

The passive monitoring process includes the following steps:

S21: Using the moving average method to eliminate the trend item of the measured signal of the piezoelectric ceramic sensor;

S22: Using the improved empirical mode decomposition method to perform signal noise reduction, and obtain a preprocessed signal near the baseline and without noise superimposition;

S23: extract the free vibration response from the signal obtained in step S22;

S24: performing modal parameter identification on the free vibration data;

S25: Establishing damage indication indicators, judging whether damage occurs and determining the degree of damage.

Step S12 specifically includes the following steps:

S12.1: Obtain the lth IMF signal c _l (t), l=1,2,3,...,;

S12.2: Subtract c _l (t) from f(t) to get r _l (t), that is, r _l (t) = f(t)-c _l (t); take r _l (t) as the original The data repeats the above steps to obtain the second component c ₂ (t) of f(t) that meets the IMF condition, and so on for n times, and obtains n components of the signal f(t) that meet the IMF condition; where, f(t ) is the deterministic component of the measured signal y(t) in step S11.

The intrinsic mode function (IMF) needs to meet two basic conditions, also known as IMF conditions:

1. The number of extreme points and zero-crossing points of the entire data segment should be the same or similar, with at most one difference.

2. The local mean is zero. At any point, the mean value of the upper envelope fitted by a finite number of local maximum points and the lower envelope fitted by a finite number of local minimum points must be zero.

Step S12.1 specifically includes the following steps:

S12.11: let k=2;

S12.12: Obtain the upper envelope u _k-1 (t) and the lower envelope v _k-1 (t) by interpolating between all local maxima through the Gaussian process;

S12.13: Calculate the mean value m _k-1 (t) of the upper and lower envelopes by formula (1):

S12.14: Subtract m _k-1 (t) from f(t) to obtain a new data sequence h _k-1 (t) with low frequency removed, namely:

h _k-1 (t)=f(t)-m _k-1 (t) (2)

Wherein, f(t) is the deterministic component of measured signal y(t) in step S11;

S12.15: Determine whether k is equal to K, K is the maximum number of iterations: if yes, end; otherwise, proceed to step S12.16;

S12.16: If h _k-1 (t) meets the IMF condition, then h _k-1 (t) is the k-1th component of f(t); otherwise, set k=k+1, and then return to the execution step S12.12.

Step S14 specifically includes the following steps:

S14.1: Extract the wavelet packet decomposition coefficient of the response signal through formula (3)

Among them, R(t) represents the new data series after denoising, ψ _j,h,i (t) is the wavelet packet with scale index j, position index h and frequency index i;

S14.2: Reconstruct the wavelet packet decomposition coefficients, extract signals in each frequency band, and obtain the wavelet packet energy spectrum E of the signal:

in, Represents the energy of the i-th frequency band, as shown in formula (5);

in, express the reconstructed signal,

In step S15, the process of identifying the degree of structural damage is as follows: select a signal transmitter to transmit a waveform signal, and identify the degree of structural damage by receiving the amplitude of the signal piezoelectric ceramic sensor: if the amplitude of the received signal is greater than the lossless If the amplitude under working conditions is smaller, it is determined that there is damage in the monitored area; if the amplitude of the received signal is equal to the amplitude under non-destructive conditions, it is determined that there is no damage in the monitored area.

In step S23, the free vibration response is extracted from the preprocessed data by random subtraction method.

The damage indication index in step S25 is the natural frequency index, and the j-th order natural frequency index f _nj is shown in formula (6):

Among them, f _uj is the j-th order natural frequency of the non-destructive structure, and f _dj is the j-th order natural frequency of the structural damage state.

The cross-sectional size of the reinforced concrete beam designed in this specific embodiment is 150mm×150mm×550mm. The piezoelectric ceramic sensor is a cylinder with a cross-sectional diameter of 30mm and a thickness of 35mm. Three piezoelectric ceramic sensors are buried in the reinforced concrete beam, and the polarization directions are all along the length direction of the reinforced concrete beam. The size of the reinforced concrete beam, the layout position of the steel bar and the embedding position of the piezoelectric ceramic sensor are shown in Figure 2.

ERPS-1 is used as the driving sensor, connected to the waveform generator during the test, and ERPS-2 and ERPS-3 are used as receivers. The active monitoring system for reinforced concrete beams is shown in Fig. 3.

First, when the reinforced concrete beam is in the undamaged state, ERPS-1 transmits the signal, and ERPS-2 and ERPS-3 receive the signal respectively. After the measurement, cut a 5mm slit in the center of the beam with a slitting machine, and use ERPS-1 to transmit signals, and ERPS-2 and ERPS-3 to receive signals respectively. Then the seam is deepened to 10mm, 15mm, ..., 30mm, and the above test is repeated. The test conditions are shown in Table 1, and the slits of reinforced concrete beams are shown in Figure 4.

Table 1 Active monitoring test of reinforced concrete beams

A 92kHz, 10Vpp sine wave is emitted by an Agilent waveform generator to collect signal data under non-destructive working conditions and six kinds of damaged working conditions. The ERPS-2 received signals under different damage conditions were compared and analyzed, and the relationship between the response signal and the damage degree was studied. Figure 5 shows the signal received by ERPS-2 when the reinforced concrete beam has no damage, the crack depth is 5 mm, and the crack depth is 10 mm. The sampling frequency of the signal is 1000 Hz. It can be seen from the figure that the signal amplitude when there is damage is smaller than that of the non-destructive condition, and the signal amplitude decreases as the depth of the crack increases. The test results show that the signal amplitude received by the piezoelectric sensor decreases with the increase of the damage degree, and the change of the signal amplitude can preliminarily diagnose the existence of structural damage and the degree of damage.

The received signals of ERPS-2 and ERPS-3 were analyzed under the same damage condition, and the relationship between the relative position between the piezoelectric sensor and the crack and the response signal was explored, as shown in Figure 6. It can be seen from Figure 6 that as the crack depth increases, the total energy index decreases, which is in line with the law that the greater the damage degree, the stress wave propagates in the structure and the energy attenuates when encountering crack damage. The total energy index of ERPS-3 is greater than that of ERPS-2, because ERPS-2 is buried closer to the fracture, and the received stress wave has a greater attenuation degree, so the total energy is smaller. Therefore, this index can be used for the preliminary location of the damage, and the crack position is on the side of the piezoelectric sensor with a smaller total energy index.

For passive monitoring, the moving average method is used to eliminate the trend item and noise reduction of the measured data collected at each measuring point, and then the processed data are respectively extracted for free vibration response processing, and then the STD method is used to identify the natural vibration frequency and shape. The natural frequency index under each working condition is calculated by using frequency and mode shape, as shown in Fig. 7.

It can be seen from Figure 7 that as the damage degree increases, the natural frequency index curves of each order generally show an upward trend, so the natural frequency index can be used to identify the damage degree of the structure.

For the judgment of damage location, it can be determined by the signal difference of different sensor monitoring parts. The propagation mode of the vibration source is to spread from the center to the surroundings. The waves generated by the vibration of the vibration source are received by the sensors arranged around it. Along the propagation direction of the wave, the damage occurrence of the structure is checked step by step. The wave received by the sensor is compared with the wave in the non-destructive state. When the wave propagates to the first sensor, if the characteristics of the wave received by the sensor change, it indicates that there is damage between the vibration source and the first sensor. , so as to complete the location of the damage; if the characteristics of the wave do not change, it means that there is no damage between the vibration source and the first sensor, and continue to verify the position of the next sensor. If the characteristics of the next wave change , it indicates that there is damage between the first sensor and the second sensor, so as to complete the location of the damage; if the characteristics of the wave do not change, it indicates that there is no damage between the first sensor and the second sensor , continue to verify the position of the next sensor until the sensor whose wave characteristics change or the last sensor is verified. For quantification of changes in wave characteristics, the amplitude index of the signal can be selected.

Claims (7)

1. A hydraulic concrete structure damage identification method based on piezoelectric ceramics is characterized in that: the method comprises an active monitoring process and a passive monitoring process, which are respectively as follows:

the active monitoring process comprises the following steps:

s11: eliminating trend items of actually measured signals y (t) of the piezoelectric ceramic sensor by using a moving average method; where t is a time signal.

S12: carrying out signal noise reduction by using an improved empirical mode decomposition method to obtain a pre-processed signal which is close to a baseline and has no noise superposition;

s13: decomposing the signal obtained in the step S12 by using a wavelet packet;

s14: solving a wavelet packet energy spectrum of the signal;

s15: establishing a damage indication index so as to identify the damage degree of the structure and position the damage;

the passive monitoring process comprises the following steps:

s21: eliminating trend terms of actually measured signals of the piezoelectric ceramic sensor by using a sliding average method;

s22: carrying out signal noise reduction by using an improved empirical mode decomposition method to obtain a pre-processed signal which is close to a baseline and has no noise superposition;

s23: extracting a free vibration response from the signal obtained in step S22;

s24: carrying out modal parameter identification on the free vibration data;

s25: and establishing a damage indication index, and judging whether damage occurs or not and determining the damage degree.

2. The hydraulic concrete structure damage identification method based on piezoelectric ceramics as claimed in claim 1, characterized in that: the step S12 specifically includes the following steps:

s12.1: obtaining the first IMF signal c_l(t)，l＝1,2,3，…，；

S12.2: subtracting c from f (t)_l(t) obtaining r_l(t), i.e. r_l(t)＝f(t)-c_l(t); will r is_l(t) repeating the above steps as raw data to obtain a second IMF-conditioned component c of f (t)₂(t), repeating the above steps for n times to obtain n IMF-conditioned components of the signal f (t); wherein f (t) is the deterministic component of the actually measured signal y (t) in step S11.

3. The hydraulic concrete structure damage identification method based on piezoelectric ceramics as claimed in claim 1, characterized in that: the step S12.1 specifically includes the steps of:

s12.11: let k be 2;

s12.12: by the Gaussian process in allInterpolation between local maxima to obtain the upper envelope u_k-1(t) and the lower envelope v_k-1(t)；

S12.13: calculating the mean value m of the upper and lower envelope lines by the formula (1)_k-1(t)：

S12.14: subtracting m from f (t)_k-1(t) finding a new data sequence h with a reduced frequency_k-1(t), namely:

h_k-1(t)＝f(t)-m_k-1(t) (2)

wherein f (t) is the deterministic component of the actually measured signal y (t) in step S11;

s12.15: judging whether K is equal to K, wherein K is the maximum iteration number: if yes, ending; otherwise, continuing to step S12.16;

s12.16: if h is_k-1(t) IMF Condition is met, then h_k-1(t) is the k-1 component of f (t); otherwise, let k be k +1, and then return to performing step S12.12.

4. The hydraulic concrete structure damage identification method based on piezoelectric ceramics as claimed in claim 1, characterized in that: the step S14 specifically includes the following steps:

s14.1: response signal wavelet packet decomposition coefficient is extracted through formula (3)

Where R (t) represents the new data series, ψ, after denoising_j,h,i(t) is a wavelet packet having a scale index j, a position index h, and a frequency index i;

s14.2: reconstructing the wavelet packet decomposition coefficient, extracting signals of each frequency band range, and solving the wavelet packet energy spectrum E of the signals:

wherein,represents the energy of the ith frequency band, as shown in equation (5);

wherein,to representThe reconstructed signal of (2).

5. The hydraulic concrete structure damage identification method based on piezoelectric ceramics as claimed in claim 1, characterized in that: in step S15, the process of identifying the structural damage degree is as follows: a signal emitter is selected to emit a waveform signal, and the structural damage degree is identified by receiving the amplitude of a signal piezoelectric ceramic sensor: if the amplitude of the received signal is smaller than that under the lossless working condition, judging that the monitored region has damage; if the amplitude of the received signal is equal to the amplitude under lossless conditions, it is determined that no damage exists in the monitored region.

6. The hydraulic concrete structure damage identification method based on piezoelectric ceramics as claimed in claim 1, characterized in that: in step S23, a free vibration response is extracted from the preprocessed data by a random subtraction method.

7. The piezoceramic-based hydraulic work of claim 1The concrete structure damage identification method is characterized by comprising the following steps: the damage indication index in step S25 is a natural frequency index, a j-th order natural frequency index f_njAs shown in formula (6):

wherein f is_ujIs the j-th order natural frequency, f, of the lossless structure_djIs the j-th order natural frequency in the state of structural damage.