Intelligent support based on acoustic emission and grating sensor and monitoring method

Abstract

The invention discloses an intelligent support based on acoustic emission and grating sensors and a monitoring method.The intelligent support comprises a support body, the upper end of the support body is connected with an upper steel plate through an annular support body limiting device, the lower end of the support body is connected with a lower steel plate through an annular support body limiting device, and two sets of baffles are arranged on the two sides of the support body; the inner sides of the two groups of baffles are respectively provided with three acoustic emission sensors, the three acoustic emission sensors are fixed on the lower steel plate through fixing devices, a plurality of mounting grooves are formed in the two ends, close to the support, of the upper steel plate and the lower steel plate, the Bragg grating sensors are arranged in the mounting grooves, and the Bragg grating sensors are electrically connected with one another. The acoustic emission sensor and the Bragg grating sensor are used in the field of bridge supports, a bridge is monitored in real time, potential problems can be found and dealt with in time, and structural damage of the bridge supports is prevented.

Description

Technical Field

[0001] The present invention relates to the technical field of bridge bearings, and in particular to an intelligent bearing and a monitoring method based on acoustic emission and grating sensors. Background Art

[0002] Bearings are commonly used in highway and railway bridges. Bearings have good shock absorption and buffering properties, which can effectively absorb and disperse bridge loads and reduce the impact of vibration and noise on the surrounding environment. The care and maintenance of bearings are also very important, and regular inspections and maintenance are required to ensure their service life and safety. If the bearings are found to be damaged or aged, they should be replaced in time to prevent adverse effects on the bridge. As the requirements for safety and reliability of infrastructure projects continue to increase, traditional bearing monitoring methods are facing increasing challenges. Traditional methods, such as mechanical sensors and visual inspections, although effective in past applications, have gradually become limited in the face of increasingly complex engineering environments and changing load conditions. These traditional methods often fail to provide sufficient real-time data, and it is difficult to fully monitor subtle changes and potential defects in bearings. Summary of the invention

[0003] In view of the deficiencies in the prior art, the present invention provides an intelligent support and monitoring method based on acoustic emission and grating sensors, which aims to solve the problems of support such as voids, stress, cracks, etc. and to provide real-time monitoring.

[0004] To achieve the above-mentioned purpose, the present invention provides the following technical solutions: an intelligent support based on acoustic emission and grating sensors, comprising an upper steel plate, a lower steel plate, a support, a ring-shaped support limit device baffle, an acoustic emission sensor and a Bragg grating sensor, wherein the upper end of the support is connected to the upper steel plate by an ring-shaped support limit device, and the lower end of the support is connected to the lower steel plate by an ring-shaped support limit device. Two groups of baffles are provided on both sides of the support, and three acoustic emission sensors are respectively provided on the inner sides of the two groups of baffles. The three acoustic emission sensors are fixed to the lower steel plate by a fixing device. Installation grooves are opened on the two ends of the upper steel plate and the lower steel plate close to the support. The Bragg grating sensor is arranged in the installation groove, and the Bragg grating sensors are electrically connected to each other.

[0005] Furthermore, the fixing device includes a gasket, a bolt and a pad. One end of the gasket is placed horizontally on the acoustic emission sensor, and the other end of the gasket is provided with a pad. The pad is welded and fixed to the lower steel plate. A reserved bolt hole is provided in the middle part of the gasket, and a bolt hole is opened on the lower steel plate near the acoustic emission sensor and the pad. The bolt passes through the reserved bolt hole of the gasket and is inserted into the bolt hole of the lower steel plate for fixation; the pad is inclined on the side close to the gasket.

[0006] Furthermore, the annular support limiting device includes an installation groove, a lead groove and an optical fiber wire. The support and the annular support limiting device are both arranged in a circular shape. An installation groove is opened on the inner side of the upper steel plate and the lower steel plate. The inner diameter of the installation groove is adapted to the outer diameter of the Bragg grating sensor.

[0007] Furthermore, wire guide grooves are provided at the upper and lower ends of the annular support limiting device, and mounting grooves are arranged at equal intervals at the four ends of the wire guide grooves, and the optical fiber conductors are installed in the wire guide grooves.

[0008] A bridge monitoring method, applied to the intelligent bearing, comprises the following steps: Step S1: lowering the bridge so that the bottom of the bridge is connected to the top of the upper steel plate of the smart support, and the bottom of the pier is connected to the lower steel plate of the smart support to complete the installation; Step S2: Detect the support by using an acoustic emission sensor to obtain an enhanced acoustic emission signal; measure the bridge and the pier by using a Bragg grating sensor to obtain an enhanced grating signal; Step S3: Processing the enhanced acoustic emission signal and the enhanced grating signal obtained in step S2 to obtain a processed enhanced acoustic emission signal and an enhanced grating signal; Step S4: using a BP neural network to fuse the processed enhanced acoustic emission signal and the enhanced grating signal to obtain a fused signal value; Step S5: Set a warning threshold, compare the signal value fused in step S4 with the warning threshold, and obtain a warning result of the bridge health status.

[0009] Further, in step S2, the enhanced acoustic emission signal is obtained, and the specific process is as follows: The support is detected by an acoustic emission sensor to obtain an acoustic emission signal, and the obtained acoustic emission signal is used as a collection It is expressed as: S = [ s 1 , s 2 , s 3 , ⋯ , s n ] (1); In the formula, Indicates the first acoustic emission signal detected; Indicates the nth acoustic emission signal detected; The energy vector carried by the acoustic emission signal is taken as a set It is expressed as: (2); In the formula, Represents the energy vector carried by the first acoustic emission signal; represents the energy vector carried by the nth acoustic emission signal; T represents transposition; The frequency vector of the acoustic emission signal is taken as a set It is expressed as: (3); In the formula, Represents the frequency vector of the first acoustic emission signal; Represents the frequency vector of the nth acoustic emission signal; The energy contribution rate of the acoustic emission signal corresponding to the energy vector carried by the acoustic emission signal and the frequency vector of the acoustic emission signal is recorded as , expressed as: (4); In the formula, Represents the energy carried by the i-th acoustic emission signal; represents the energy carried by the kth acoustic emission signal; Represents the frequency vector of the kth acoustic emission signal; The cumulative energy contribution rate of the acoustic emission signal is calculated based on the energy vector carried by the acoustic emission signal and the frequency vector of the acoustic emission signal. It is expressed as: (5); The threshold is set to the average energy value of the acoustic emission signal between 85% and 95% according to the cumulative energy contribution rate; It is expressed as: (6); In the formula, max (E) represents the maximum energy value of the acoustic emission signal; Q represents the preliminary threshold; R represents the proportion of the number of signals whose acoustic emission signal energy is greater than the preliminary optimal threshold; Threshold-based Select the energy value of the acoustic emission signal to construct the key characteristic signal set of the acoustic emission signal precursor , expressed as: (7); In the formula, represents the i-th acoustic emission signal detected; E(i) represents the energy value of the i-th acoustic emission signal; According to the final acoustic emission signal precursor key characteristic signal set Obtain enhanced acoustic emission signals.

[0010] Further, in step S2, the enhanced grating signal is obtained, and the specific process is as follows: The Bragg grating sensor detects external light waves, and the light waves that meet the matching conditions of the Bragg grating sensor will be strongly reflected, and the light waves that do not meet the matching conditions of the Bragg grating sensor will be weakly reflected; The condition of the Bragg grating sensor is expressed as: (8); In the formula, It represents the central wavelength of the incident light wave reflected back by the Bragg grating sensor in free space; It represents the refractive index of the Bragg grating sensor for the central wavelength in free space; Indicates the spacing length between Bragg grating sensors; When the measured physical quantity of the bridge or pier changes, the Bragg grating sensor and Λ will change, and partial differentiation of equation (8) yields and Affects the central wavelength drift The table is: (9); In the formula, It represents the change of the refractive index of the Bragg grating sensor with respect to the central wavelength of free space; Indicates the change in the spacing length between Bragg grating sensors; The central wavelength shift of the Bragg grating sensor reflection Wavelength demodulation is performed, and the Bragg grating sensor after wavelength demodulation is used to measure the strain of bridges and piers to obtain enhanced grating signals; wavelength drift The wavelength demodulation is expressed as: Δ λ B = [ ( 1 − p e ) Δ ε + ( ξ + α ) Δ T ] λ (10); In the formula, represents the strain optical sensitivity coefficient; Indicates the strain effect on the Bragg grating sensor; represents the coefficient of thermal expansion; represents the temperature optical sensitivity coefficient; represents the temperature effect on the Bragg grating sensor; the enhanced grating signal obtained is expressed as a set M t = [ m t 1 , m t 2 , m t 3 , ⋯ , m t n ] ; Indicates the first enhanced grating signal; Represents the nth enhanced grating signal.

[0011] Further, in step S3, the processed enhanced acoustic emission signal and enhanced grating signal are obtained, and the specific process is as follows: Step S31: Processing the enhanced acoustic emission signal and the enhanced grating signal respectively, removing the random noise and environmental noise contained in the enhanced grating signal; removing the abnormal part of the enhanced acoustic emission signal caused by the external environment interference, and obtaining the processed enhanced acoustic emission signal and enhanced grating signal; Step S311: removing random noise and environmental noise contained in the enhanced grating signal, the process is as follows: The collected The enhanced grating signals are divided into N groups, and the five-point cubic polynomial center moving smoothing method is used on the N groups to remove the random noise of the enhanced grating signals, which is expressed as: y 1 = 1 70 [ 69 x 1 + 4 ( x 2 + x 4 ) − 6 x 3 − x 5 ] y 2 = 1 35 [ 27 x 1 + 2 ( x 1 + x 5 ) + 12 x 3 − 8 x 4 ] y 3 = 1 35 [ − 3 ( x 1 + x 5 ) + 12 ( x 2 + x 4 ) + 17 x 3 ] y 4 = 1 35 [ 2 ( x 1 + x 5 ) + 27 x 4 + 12 x 3 − 8 x 2 ] y 5 = 1 70 [ 4 ( x 2 + x 4 ) + 69 x 5 − 6 x 3 − x 1 ] (11); In the formula, represents the i-th data point after the five-point cubic polynomial center moving smoothing process; represents the i-th data point in the enhanced grating signal; i=1,2,3,…,5; Step S312: using a linear regression analysis method to study the relationship between the environmental noise factor and the support factor, and removing the environmental noise in the enhanced grating signal; the linear regression analysis method includes a univariate linear regression analysis method and a multivariate linear regression analysis method; The univariate linear regression analysis method is expressed as: (12); In the formula, represents the dependent variables related to ambient noise affected by temperature load and wind speed load; and All represent regression coefficients; Represents the temperature load and wind speed load noise factors in the environmental noise; represents the random error term; Assumptions , then the random variable ; represents normal distribution; represents the standard deviation of the random error term; The multiple linear regression analysis method is expressed as: (13); In the formula, represents the regression coefficient; Represents the temperature load noise factor in the ambient noise; Step S313: removing the abnormal part of the enhanced acoustic emission signal caused by external environmental interference, the process is as follows: The obtained enhanced acoustic emission signal contains ; represents the nth enhanced acoustic emission signal; Enhanced AE signal average It is expressed as: (14); In the formula, represents the i-th enhanced acoustic emission signal; Enhanced acoustic emission signal residual error It is expressed as: (15); According to the Saibel formula, the standard deviation of the enhanced acoustic emission signal is obtained It is expressed as: (16); With residual error and standard deviation The ratio is judged, and the ratio is expressed as: (17); when , which means that the enhanced acoustic emission signal detected is interfered by the external environment, and the abnormal part of the enhanced acoustic emission signal caused by the external environment interference is removed; It is the ratio of the residual error to the standard deviation.

[0012] Further, in step S4, the fused signal value is obtained, and the specific process is as follows: The BP neural network is used to fuse the processed enhanced acoustic emission signal and enhanced grating signal to obtain the fused signal value; the specific process is as follows: A three-layer BP neural network is established and trained; the BP neural network consists of an input layer, a hidden layer, and an output layer. As the transfer function, we use As the error back propagation function; where represents the transfer function in the BP neural network; e represents the exponential function; Represents the expected output, and continuously adjusts the weights and thresholds of the BP neural network to minimize the error back propagation function E; Represents the calculation output of the BP neural network; The processed enhanced acoustic emission signal and enhanced grating signal are input into the trained BP neural network for fusion to obtain the fused single environmental factor data; The fused single environmental factor data are fused at the feature level to obtain the fused signal value.

[0013] Further, in step S5, the early warning result of the bridge health status is obtained, and the specific process is as follows: Step S51: Compare the fused signal value in step S4 with the warning thresholds of the three conditions of "normal, warning, and dangerous", and finally obtain the warning result of the health status of the bridge. The specific process is as follows: Step S511: sorting out the historical data of the strong acoustic emission signal and the enhanced grating signal, removing the anomalies in the historical data, and calculating the change rate of each point in the historical data as a sample; Calculate sample expected value It is expressed as: (18); In the formula, Indicates the deformation rate of the monitoring point; In formula (11) Select r data as a group, select m groups in total, and get the data ; represents the deformation rate of the i-th monitoring point in the j-th group of data; Indicates the number of data in each data group; Indicates the sequence number of the data group; Indicates the data group selected from the overall data; Calculate the expected value of m groups of data ; represents the deformation rate in the jth group of data, calculated from the mth group of data It is expressed as: (19); (20); (twenty one); (twenty two); (twenty three); (twenty four); (25); In the formula, represents the average value of m groups of data; Indicates the mth variance of the new data calculated from m groups of data; It represents the sample expected value of m groups of new data after calculating the variance of new data from m groups of data; Indicates the jth variance of the new data calculated from m groups of data; Represents each data point in the jth group of data The difference from the mean value of the group of data The sum of the fourth power of is divided by r-1; represents sample expectation; Indicates the variance of the new data calculated from m groups of data; It represents the sample variance of the new data after the new data composition variance is calculated from m groups of data; according to and Obtain the entropy value of the cloud model and super entropy It is expressed as; (26); (27); According to sample expectations , model entropy and super entropy Obtain the cloud droplet diagram; The bridge safety operation value is set according to the cloud droplet diagram, and the deformation rate of the bridge safety operation value is between; According to the bridge safety operation value, the warning thresholds of the three conditions of "normal, warning, and dangerous" are set. The warning thresholds of the three conditions of "normal, warning, and dangerous" are set to 1.3 times, 1.5 times, and 2 times of the safe state respectively; The real-time monitoring fusion signal value is compared with the warning thresholds of the three conditions of "normal, warning, and dangerous" to obtain the warning results of the bridge health status.

[0014] Compared with the existing technology, the present invention has the following beneficial effects: (1) The present invention uses acoustic emission sensors and Bragg grating sensors in the field of bridge bearings, and adopts BP neural network to integrate and optimize the data of two heterogeneous sensors, so that the bridge bearings can identify early signals of crack propagation and fatigue damage in the material during service; it can also monitor in real time whether the bridge bearings are empty or offset. Real-time monitoring helps to timely discover and respond to potential problems and prevent the occurrence of structural damage to the bridge bearings.

[0015] (2) The present invention forms a fixing device for the acoustic emission sensor by combining gaskets, bolts, pads and other components. Compared with the traditional fixing method, it is more ingenious and simpler, and its stability performance is greatly improved.

[0016] (3) The present invention pre-processes the acoustic emission signals and grating signals detected by the acoustic emission sensor and the Bragg grating sensor, sets the safe operation value of the bridge, and sets the early warning threshold. The safe operation value monitored in real time is compared with the early warning threshold, and early warning information is issued to achieve real-time monitoring of the health status of the bridge, which can effectively evaluate the condition of the bridge during operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a schematic diagram of the three-dimensional structure of the intelligent support of the acoustic emission sensor and grating sensor of the present invention.

[0018] Figure 2 It is a schematic diagram of the planar structure of the intelligent support of the acoustic emission sensor and grating sensor of the present invention.

[0019] Figure 3 It is a cross-sectional schematic diagram of the intelligent support of the acoustic emission sensor and grating sensor of the present invention.

[0020] Figure 4 It is a schematic diagram of a fixed plane of the acoustic emission sensor of the present invention.

[0021] Figure 5 It is a plan view of the mounting groove and the wire guide groove of the present invention.

[0022] Figure 6 The present invention is a flow chart of the method.

[0023] Markings in the figure: 1. upper steel plate; 2. lower steel plate; 3. support; 4. annular support limit device; 5. baffle; 6. gasket; 7. acoustic emission sensor; 8. bolt; 9. pad; 10. mounting groove; 11. Bragg grating sensor; 12. lead groove; 13. optical fiber conductor. DETAILED DESCRIPTION

[0024] like Figure 1-Figure 3 As shown, the present invention provides a technical solution: an intelligent support based on acoustic emission and grating sensors, comprising an upper steel plate 1, a lower steel plate 2, a support 3, an annular support limiting device 4, a baffle 5, an acoustic emission sensor 7 and a Bragg grating sensor 11; the upper end of the support 3 is connected to the upper steel plate 1 through the annular support limiting device 4, and the lower end of the support 3 is connected to the lower steel plate 2 through the annular support limiting device 4, and two groups of baffles 5 are provided on both sides of the support 3; three acoustic emission sensors 7 are respectively arranged on the inner side of each group of baffles 5, and the three acoustic emission sensors 7 are fixed on the lower steel plate 2 by a fixing device, and mounting grooves 10 are opened on both ends of the upper steel plate 1 and the lower steel plate 2 close to the support 3, and the Bragg grating sensors 11 are arranged in the mounting grooves 10, and the Bragg grating sensors 11 are electrically connected to each other.

[0025] Among them, the fixing device of the acoustic emission sensor 7 includes a gasket 6, a bolt 8 and a pad 9. When the quiet air condition is met, one end of the gasket 6 is placed horizontally on the acoustic emission sensor 7, and the other end of the gasket 6 is provided with a pad 9. The pad 9 is pushed horizontally until it contacts the other end of the gasket 6, and the pad 9 is welded and fixed to the lower steel plate 2. A reserved bolt hole is provided in the middle part of the gasket 6, and a bolt hole is opened on the lower steel plate 2 near the acoustic emission sensor 7 and the pad 9. The bolt 8 is inserted into the bolt hole of the lower steel plate 2 through the reserved bolt hole of the gasket 6 and tightened, thereby realizing the fixation of the acoustic emission sensor 7.

[0026] Among them, the side of the cushion block 9 close to the gasket 6 is an inclined surface, and the gasket 6 is pressed onto the acoustic emission sensor 7 and the cushion block 9 by the bolt 8, so as to fix the acoustic emission sensor 7; the acoustic emission sensor 7 is required to be in the front third of the cushion block 9, and the acoustic emission sensor 7 is at a certain distance from the front end of the cushion block 9.

[0027] like Figure 4As shown, the main functions of the upper steel plate 1 and the lower steel plate 2 are to increase the contact area between the support 3 and the upper beam and the lower bridge pier to reduce stress concentration, and to fix the Bragg grating sensor 11 through the installation groove 10, and fix the support 3 between the upper steel plate 1 and the lower steel plate 2 through the annular support limit device 4; the support 3 adopts a rubber support, and its main function is to support and buffer the beam of the bridge; the Bragg grating sensor 11 mainly measures the deformation and stress distribution of the support 3 under the longitudinal load, and the two groups of Bragg grating sensors 11 can sense each other, and determine whether there is a risk of air loss by measuring the relative displacement between the upper steel plate 1 and the lower steel plate 2; at the same time, the acoustic emission sensor 7 can be quickly replaced by unscrewing the bolt 8 and removing the gasket 6. By using the acoustic emission AE technology and the Bragg grating principle to monitor the health status of the bridge in real time, potential problems of the bridge can be predicted in advance, so that it can reach the normal service life and increase its service life.

[0028] like Figure 5 As shown, the annular support limiting device 4 includes a mounting groove 10, a lead groove 12 and an optical fiber conductor 13. The support 3 and the annular support limiting device 4 are both arranged in a circular shape. The main function of the annular support limiting device 4 is to connect one end of the upper steel plate 1 and the lower steel plate 2 to form an annular limit. The upper steel plate 1 and the lower steel plate 2 are provided with a mounting groove 10 on the inner side, and the inner diameter of the mounting groove 10 is adapted to the outer diameter of the Bragg grating sensor 11. The mounting groove 10 is mainly used to install the Bragg grating sensor 11. When the Bragg grating sensor 11 needs to be replaced, the support 3 only needs to be pulled out of the mounting groove 10 and a new Bragg grating sensor 11 needs to be inserted.

[0029] Among them, the upper and lower ends of the annular support limiting device 4 are provided with lead grooves 12, the mounting grooves 10 are arranged at equal intervals at the four ends of the lead grooves 12, the optical fiber wires 13 are installed in the lead grooves 12, and the Bragg grating sensors 11 are connected in series through the optical fiber wires 13 and sense each other. The mounting grooves 10 on the lead grooves 12 are arranged at equal intervals at the four sides of the upper steel plate 1 and the lower steel plate 2, and a total of four groups are set, that is, one steel plate can be installed with four groups of Bragg grating sensors 11.

[0030] Among them, the function of the lead groove 12 is to accommodate the optical fiber wire 13 so that the optical fiber wire 13 does not directly contact the support 3, the upper steel plate 1, and the lower steel plate 2, thereby avoiding the optical fiber wire 13 from being damaged under the longitudinal vehicle load between the upper steel plate 1 and the support 3.

[0031] The depth of the mounting groove 10 is three quarters of the Bragg grating sensor 11 , and the inner diameter of the mounting groove 10 is two thirds of the Bragg grating sensor 11 .

[0032] The installation steps of the smart support include: Insert a plurality of Bragg grating sensors 11 into the mounting grooves 10 on the upper steel plate 1 and the lower steel plate 2, respectively, and make the four groups of Bragg grating sensors 11 on the support 3 electrically connected to each other, install the four groups of Bragg grating sensors 11 in the corresponding mounting grooves 10, and connect the four groups of Bragg grating sensors 11 in series through the optical fiber wire 13, and finally fix the optical fiber wire 13 in the lead groove 12; use the annular support limit device 4 to limit the upper steel plate 1, the lower steel plate 2 and the support 3, and under the condition of meeting the static air condition, arrange the acoustic emission sensors 7 at equal intervals on both sides outside the circumference of the support 3, each There are three on the side, with a total of six acoustic emission sensors 7. After the acoustic emission sensor 7 is bonded to the lower steel plate 2 by a strong structural colloid, one end of the gasket 6 is horizontally placed on the acoustic emission sensor 7, and after the pad 9 is horizontally moved toward one end of the acoustic emission sensor 7 and contacts with the other end of the pad 6, the pad 9 is welded to the lower steel plate 2, and the bolts 8 are inserted into the bolt holes of the lower steel plate 2 through the reserved bolt holes of the gasket 6 and tightened. Finally, the baffle 5 is welded and connected to the upper steel plate 1 and the lower steel plate 2, and the beam body is lowered to connect the bottom of the beam body with the top of the upper steel plate 1, and the bottom of the pier is connected to the lower steel plate 2 to complete the installation.

[0033] like Figure 6 As shown, a bridge monitoring method, applied to the intelligent bearing, comprises the following steps: Step S1: lowering the bridge so that the bottom of the bridge is connected to the top of the upper steel plate of the smart support, and the bottom of the pier is connected to the lower steel plate of the smart support to complete the installation; Step S2: Detect the support by using an acoustic emission sensor to obtain an enhanced acoustic emission signal; measure the bridge and the pier by using a Bragg grating sensor to obtain an enhanced grating signal; Step S3: Processing the enhanced acoustic emission signal and the enhanced grating signal obtained in step S2 to obtain a processed enhanced acoustic emission signal and an enhanced grating signal; Step S4: using a BP neural network to fuse the processed enhanced acoustic emission signal and the enhanced grating signal to obtain a fused signal value; Step S5: Set a warning threshold, compare the signal value fused in step S4 with the warning threshold, and obtain a warning result of the bridge health status.

[0034] Among them, the enhanced acoustic emission signal is obtained in step S2, and the specific process is: The support is detected by an acoustic emission sensor to obtain an acoustic emission signal, and the obtained acoustic emission signal is used as a collection It is expressed as: S = [ s 1 , s 2 , s 3 , ⋯ , s n ] (1); In the formula, Indicates the first acoustic emission signal detected; Indicates the nth acoustic emission signal detected; The collection of energy vectors carried by the acoustic emission signal It is expressed as: (2); In the formula, Indicates the energy carried by the first acoustic emission signal; represents the energy carried by the nth acoustic emission signal; T represents transposition; The collection of frequency vectors of acoustic emission signals It is expressed as: (3); In the formula, Represents the frequency vector of the first acoustic emission signal; Represents the frequency vector of the nth acoustic emission signal; The energy contribution rate of the acoustic emission signal corresponding to the energy vector carried by the acoustic emission signal and the frequency vector of the acoustic emission signal is recorded as , expressed as: (4); In the formula, Represents the energy carried by the i-th acoustic emission signal; represents the energy carried by the kth acoustic emission signal; Represents the frequency vector of the kth acoustic emission signal; The cumulative energy contribution rate of the acoustic emission signal is calculated based on the energy vector carried by the acoustic emission signal and the frequency vector of the acoustic emission signal. It is expressed as: (5); The threshold is set to the average energy value of the acoustic emission signal between 85% and 95% according to the cumulative energy contribution rate; It is expressed as: (6); In the formula, max (E) represents the maximum energy value of the acoustic emission signal; Q represents the preliminary threshold (the average value of the signal energy); R represents the proportion of the number of signals whose acoustic emission signal energy is greater than the preliminary optimal threshold; Threshold-based Select the energy value of the acoustic emission signal to construct the key characteristic signal set of the acoustic emission signal precursor , expressed as: (7); In the formula, represents the i-th acoustic emission signal detected; E(i) represents the energy value of the i-th acoustic emission signal; According to the final acoustic emission signal precursor key characteristic signal set Obtain enhanced acoustic emission signals.

[0035] Among them, the acoustic emission sensor (AE) detection technology is applied to the direction of intelligent bridge bearings; when the concrete structure undergoes plastic deformation or fracture due to stress, its internal microstructure will change, and energy will be released from the structure to the outside in the form of instantaneous elastic waves (AE source); the released energy changes the dynamic strain field around the AE source and generates mechanical disturbances, called AE original waves; the original AE waves propagate along the solid medium and cause mechanical vibrations on the surface of the medium; the acoustic emission sensor installed on the surface of the medium can convert the received mechanical disturbances into acoustic emission signals; after recording, amplifying and processing the emission signals, the damage inside the concrete structure can be inferred through later analysis.

[0036] Acoustic emission sensor (AE) sensor selected is SAEU3H acoustic emission detector, which has a total of 6 acquisition cards, each with 4 independent channels, a single-channel maximum sampling rate of 10M points / second, and a response frequency of 1kHz~2.5MHz; it is characterized by a unique waveform stream data storage function, which can continuously store acoustic emission signals to the hard disk at a rate of 10M sampling points per second; the acoustic emission signal on each channel is processed in high-speed real-time through the FPGA hardware of the hard disk, and can collect and analyze the processed acoustic emission signal data in real time, with highly accurate signal processing capabilities.

[0037] Among them, the enhanced grating signal is obtained in step S2, and the specific process is: When the external light waves pass through the Bragg grating sensor 11, the Bragg grating sensor 11 detects the external light waves, and forms a strong reflection for the light waves that meet the matching conditions of the Bragg grating sensor, and forms a weak reflection for the light waves that do not meet the matching conditions of the Bragg grating sensor. The Bragg grating sensor 11 is equivalent to a reflector. The incident light waves that meet the conditions of the Bragg grating sensor 11 will be reflected back, while the light waves that do not meet this condition will continue to move forward along the optical path; The condition of the Bragg grating sensor 11 is expressed as: (8); In the formula, It represents the central wavelength of the incident light wave reflected back by the Bragg grating sensor in free space; It represents the refractive index of the Bragg grating sensor for the central wavelength in free space; Indicates the spacing length between Bragg grating sensors; When the measured physical quantity of the bridge or pier changes, the Bragg grating sensor 11 and Λ will change; from formula (8), it can be seen that the central wavelength of the Bragg grating sensor 11 will also produce a corresponding drift; partial differentiation calculation of formula (8) yields and Affects the central wavelength drift The table is: (9); In the formula, It represents the change of the refractive index of the Bragg grating sensor with respect to the central wavelength of free space; Indicates the change in the spacing length between Bragg grating sensors; The central wavelength shift of the Bragg grating sensor 11 Perform wavelength demodulation; The nature of using the Bragg grating sensor 11 after wavelength demodulation to measure the strain of bridges and piers is that changing the Bragg grating sensor 11 distance causes the central wavelength to shift. Temperature and strain will cause it to change. When measuring a single variable of a bridge or pier, it is necessary to eliminate the influence of another variable as much as possible to measure an accurate value. The strain of bridges and piers is mainly measured to obtain enhanced grating signals. The wavelength drift formula is expressed as: Δ λ B = [ ( 1 − p e ) Δ ε + ( ξ + α ) Δ T ] λ (10); In the formula, represents the strain optical sensitivity coefficient; Indicates the strain effect on the Bragg grating sensor; represents the coefficient of thermal expansion; represents the temperature optical sensitivity coefficient; represents the temperature effect on the Bragg grating sensor; the enhanced grating signal obtained is expressed as a set M t = [ m t 1 , m t 2 , m t 3 , ⋯ , m t n ] ; Indicates the first enhanced grating signal; represents the nth enhanced grating signal; The characteristics of Bragg grating sensors are used to measure and monitor changes in physical quantities. Bragg grating sensors are a subclass of fiber optic sensors that use Bragg grating sensors in optical fiber conductors to achieve frequency-selective reflection of light waves, thereby monitoring the strain of bridges and piers. Compared with traditional Bragg grating sensors, they have the characteristics of high sensitivity, strong anti-interference ability, long ranging capability, multi-parameter measurement, strong adaptability to complex and extreme conditions, no power requirement, and light weight.

[0038] Among them, in step S3, the processed enhanced acoustic emission signal and enhanced grating signal are obtained, and the specific process is: Step S31: Processing the enhanced acoustic emission signal and the enhanced grating signal respectively, removing random noise and environmental noise contained in the enhanced grating signal; removing the abnormal part of the enhanced acoustic emission signal caused by external environmental interference, and obtaining the processed enhanced acoustic emission signal and enhanced grating signal; Step S311: removing random noise and environmental noise contained in the enhanced grating signal, the process is as follows: For random noise, the purpose of smoothing the enhanced grating signal detected is to obtain a set of data with more noise eliminated or reduced after processing. The enhanced grating signals are divided into N groups, and the five-point cubic polynomial center moving smoothing method is used on the N groups to remove the random noise of the enhanced grating signals, which is expressed as: y 1 = 1 70 [ 69 x 1 + 4 ( x 2 + x 4 ) − 6 x 3 − x 5 ] y 2 = 1 35 [ 27 x 1 + 2 ( x 1 + x 5 ) + 12 x 3 − 8 x 4 ] y 3 = 1 35 [ − 3 ( x 1 + x 5 ) + 12 ( x 2 + x 4 ) + 17 x 3 ] y 4 = 1 35 [ 2 ( x 1 + x 5 ) + 27 x 4 + 12 x 3 − 8 x 2 ] y 5 = 1 70 [ 4 ( x 2 + x 4 ) + 69 x 5 − 6 x 3 − x 1 ] (11); In the formula, represents the i-th data point after the five-point cubic polynomial center moving smoothing process; represents the i-th data point in the enhanced grating signal; i=1,2,3,…,5; Step S312: The principle of using the five-point cubic polynomial center moving smoothing method is the least square method, which can process the random noise of the enhanced grating signal and reduce the distortion of the data; Step S313: For environmental noise, the environmental noise clock is mainly affected by temperature load and wind speed load. The linear regression analysis method can be used to study the relationship between the environmental noise factor and the support factor to remove the environmental noise in the enhanced grating signal; the linear regression analysis method includes univariate linear regression analysis method and multivariate linear regression analysis method; The univariate linear regression analysis method is expressed as: (12); In the formula, represents the dependent variables related to ambient noise affected by temperature load and wind speed load; and All represent regression coefficients; Represents the temperature load and wind speed load noise factors in the environmental noise; represents the random error term; Assumptions , then the random variable ;in, represents normal distribution; It represents the standard deviation of the random error term, which reflects the degree of dispersion of the random error term; The multiple linear regression analysis method is expressed as: (13); In the formula, represents the regression coefficient; Represents the temperature load noise factor in the ambient noise; Step S314: removing the abnormality of the emission signal caused by the external environment contained in the enhanced acoustic emission signal, the process is as follows: Step S315: During bridge monitoring, due to interference from the external environment, the detected enhanced acoustic emission signal may be abnormal; the Laida rule is used to remove the abnormal part of the enhanced acoustic emission signal caused by interference from the external environment; Step S316: The obtained enhanced acoustic emission signal includes ; represents the nth enhanced acoustic emission signal; Enhanced AE signal average It is expressed as: (14); In the formula, represents the i-th enhanced acoustic emission signal; Enhanced acoustic emission signal residual error It is expressed as: (15); According to the Saibel formula, the standard deviation of the enhanced acoustic emission signal is obtained It is expressed as: (16); With residual error and standard deviation The ratio is judged, and the ratio is expressed as: (17); when , which means that the enhanced acoustic emission signal detected is interfered by the external environment, and the abnormal part of the enhanced acoustic emission signal caused by the external environment interference is removed; It is the ratio of the residual error to the standard deviation.

[0039] Among them, the fused signal value is obtained in step S4, and the specific process is: The BP neural network is used to fuse the processed enhanced acoustic emission signal and enhanced grating signal to obtain the fused signal value; the specific process is as follows: A three-layer BP neural network is established and trained; the BP neural network consists of an input layer, a hidden layer, and an output layer. As the transfer function, we use As the error back propagation function; where represents the transfer function in the BP neural network; e represents the exponential function; Represents the expected output, and continuously adjusts the weights and thresholds of the BP neural network to minimize the error back propagation function E; Represents the calculation output of the BP neural network; The processed enhanced acoustic emission signal and enhanced grating signal are input into the trained BP neural network for fusion to obtain the fused single environmental factor data; The fused single environmental factor data are fused at the feature level to obtain the fused signal value.

[0040] Among them, the early warning result of the bridge health status is obtained in step S5, and the specific process is: Step S51: Compare the fused signal value in step S4 with the warning thresholds of the three conditions of "normal, warning, and dangerous" to obtain the warning result of the bridge health status, and provide real-time feedback on the warning result. The specific process is as follows: The historical data of strong acoustic emission signals and enhanced grating signals are sorted out to remove the anomalies in the historical data, and the change rate of each point in the historical data is calculated as a sample; Calculate sample expected value It is expressed as: (18); In the formula, Indicates the deformation rate of the monitoring point; In formula (11) Select r data as a group, select m groups in total, and get the data ; represents the deformation rate of the i-th monitoring point in the j-th group of data; Indicates the sequence number of the monitoring point in the data group; Indicates the number of data in each data group; Indicates the serial number of the data group, in order to distinguish it from i; Indicates the data group selected from the overall data; Calculate the expected value of m groups of data ; represents the deformation rate in the jth group of data, calculated from the mth group of data It is expressed as: (19); (20); (twenty one); (twenty two); (twenty three); (twenty four); (25); In the formula, represents the average value of m groups of data; Indicates the mth variance of the new data calculated from m groups of data; It represents the sample expected value of the new data after the variance of the new data is calculated from the m groups of data; Indicates the jth variance of the new data calculated from m groups of data; Represents each data point in the jth group of data The difference from the mean value of the group of data The sum of the fourth power of is divided by r-1; represents sample expectation; Indicates the variance of the new data calculated from m groups of data; It means the sample variance of m groups of new data after calculating the variance of new data from m groups of data; according to and Obtain the entropy value of the cloud model and super entropy It is expressed as; (26); (27); According to sample expectations , model entropy and super entropy Obtain the cloud droplet diagram; The bridge safety operation value is set according to the cloud droplet diagram, and its deformation rate is between; According to the bridge safety operation value, the warning thresholds of the three conditions of "normal, warning, and dangerous" are set. The warning thresholds of the three conditions of "normal, warning, and dangerous" are set to 1.3 times, 1.5 times, and 2 times of the safe state respectively; The real-time monitoring fusion signal value is compared with the warning thresholds of the three conditions of "normal, warning, and dangerous", and finally the warning result of the bridge health status is obtained and the warning information is issued.

[0041] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent support based on acoustic emission and grating sensors, comprising an upper steel plate, a lower steel plate, a support, an annular support limit device, a baffle, an acoustic emission sensor and a Bragg grating sensor, characterized in that: The upper end of the support is connected to the upper steel plate by an annular support limiting device, and the lower end of the support is connected to the lower steel plate by an annular support limiting device. Two groups of baffles are provided on both sides of the support, and three acoustic emission sensors are respectively provided on the inner sides of the two groups of baffles. The three acoustic emission sensors are fixed on the lower steel plate by a fixing device. Installation grooves are opened on both ends of the upper steel plate and the lower steel plate close to the support. Bragg grating sensors are arranged in the installation grooves, and the Bragg grating sensors are electrically connected to each other.

2. According to claim 1, the intelligent support based on acoustic emission and grating sensor is characterized in that: The fixing device comprises a gasket, a bolt and a pad. One end of the gasket is horizontally placed on the acoustic emission sensor, and the other end of the gasket is provided with a pad. The pad is welded and fixed to the lower steel plate. A reserved bolt hole is provided in the middle part of the gasket. A bolt hole is opened on the lower steel plate near the acoustic emission sensor and the pad. The bolt passes through the reserved bolt hole of the gasket and is inserted into the bolt hole of the lower steel plate for fixation. The pad is inclined on the side close to the gasket.

3. The intelligent support based on acoustic emission and grating sensor according to claim 2 is characterized in that: The annular support limiting device includes an installation groove, a lead groove and an optical fiber wire. The support and the annular support limiting device are both arranged in a circular shape. An installation groove is opened on the inner side of the upper steel plate and the lower steel plate. The inner diameter of the installation groove is adapted to the outer diameter of the Bragg grating sensor.

4. The intelligent support based on acoustic emission and grating sensor according to claim 3 is characterized in that: Lead-in grooves are arranged at the upper and lower ends of the annular support limiting device, and mounting grooves are arranged at four ends of the lead-in grooves at equal intervals, and the optical fiber conductors are installed in the lead-in grooves.

5. A bridge monitoring method, applied to the smart bearing according to any one of claims 1 to 4, characterized in that: The following steps are involved: Step S1: lowering the bridge so that the bottom of the bridge is connected to the top of the upper steel plate of the smart support, and the bottom of the pier is connected to the lower steel plate of the smart support to complete the installation; Step S2: Detect the support by using an acoustic emission sensor to obtain an enhanced acoustic emission signal; measure the bridge and the pier by using a Bragg grating sensor to obtain an enhanced grating signal; Step S3: Processing the enhanced acoustic emission signal and the enhanced grating signal obtained in step S2 to obtain a processed enhanced acoustic emission signal and an enhanced grating signal; Step S4: using a BP neural network to fuse the processed enhanced acoustic emission signal and the enhanced grating signal to obtain a fused signal value; Step S5: Set a warning threshold, compare the signal value fused in step S4 with the warning threshold, and obtain a warning result of the bridge health status.

6. A bridge monitoring method according to claim 5, characterized in that: In step S2, the enhanced acoustic emission signal is obtained, and the specific process is as follows: The support is detected by an acoustic emission sensor to obtain an acoustic emission signal, and the obtained acoustic emission signal is used as a collection It is expressed as: （1）； In the formula, Indicates the first acoustic emission signal detected; Indicates the nth acoustic emission signal detected; The energy vector carried by the acoustic emission signal is taken as a set It is expressed as: （2）； In the formula, Represents the energy vector carried by the first acoustic emission signal; represents the energy vector carried by the nth acoustic emission signal; T represents transposition; The frequency vector of the acoustic emission signal is taken as a set It is expressed as: （3）； In the formula, Represents the frequency vector of the first acoustic emission signal; Represents the frequency vector of the nth acoustic emission signal; The energy contribution rate of the acoustic emission signal corresponding to the energy vector carried by the acoustic emission signal and the frequency vector of the acoustic emission signal is recorded as , expressed as: （4）； In the formula, Represents the energy carried by the i-th acoustic emission signal; represents the energy carried by the kth acoustic emission signal; Represents the frequency vector of the kth acoustic emission signal; The cumulative energy contribution rate of the acoustic emission signal is calculated based on the energy vector carried by the acoustic emission signal and the frequency vector of the acoustic emission signal. It is expressed as: （5）； The threshold is set to the average energy value of the acoustic emission signal between 85% and 95% according to the cumulative energy contribution rate; It is expressed as: （6）； In the formula, max (E) represents the maximum energy value of the acoustic emission signal; Q represents the preliminary threshold; R represents the proportion of signals whose acoustic emission signal energy is greater than the preliminary optimal threshold; Threshold-based Select the energy value of the acoustic emission signal to construct the key characteristic signal set of the acoustic emission signal precursor , expressed as: （7）； In the formula, represents the i-th acoustic emission signal detected; E(i) represents the energy value of the i-th acoustic emission signal; According to the final acoustic emission signal precursor key characteristic signal set Obtain enhanced acoustic emission signals.

7. A bridge monitoring method according to claim 6, characterized in that: In step S2, the enhanced grating signal is obtained, and the specific process is as follows: The Bragg grating sensor detects external light waves, and the light waves that meet the matching conditions of the Bragg grating sensor will be strongly reflected, and the light waves that do not meet the matching conditions of the Bragg grating sensor will be weakly reflected; The condition of the Bragg grating sensor is expressed as: （8）； In the formula, It represents the central wavelength of the incident light wave reflected back by the Bragg grating sensor in free space; It represents the refractive index of the Bragg grating sensor for the central wavelength in free space; Indicates the spacing length between Bragg grating sensors; When the measured physical quantity of the bridge or pier changes, the Bragg grating sensor and Λ will change, and partial differentiation of equation (8) yields and Affects the central wavelength drift The table is: （9）； In the formula, It represents the change of the refractive index of the Bragg grating sensor with respect to the central wavelength of free space; Indicates the change in the spacing length between Bragg grating sensors; The central wavelength shift of the Bragg grating sensor reflection Wavelength demodulation is performed, and the Bragg grating sensor after wavelength demodulation is used to measure the strain of bridges and piers to obtain enhanced grating signals; wavelength drift The wavelength demodulation is expressed as: （10）； In the formula, represents the strain optical sensitivity coefficient; Indicates the strain effect on the Bragg grating sensor; represents the coefficient of thermal expansion; represents the temperature optical sensitivity coefficient; represents the temperature effect on the Bragg grating sensor; the enhanced grating signal obtained is expressed as a set ; Indicates the first enhanced grating signal; Represents the nth enhanced grating signal.

8. A bridge monitoring method according to claim 7, characterized in that: In step S3, the processed enhanced acoustic emission signal and enhanced grating signal are obtained, and the specific process is as follows: Step S31: Processing the enhanced acoustic emission signal and the enhanced grating signal respectively, removing the random noise and environmental noise contained in the enhanced grating signal; removing the abnormal part of the enhanced acoustic emission signal caused by the external environment interference, and obtaining the processed enhanced acoustic emission signal and enhanced grating signal; Step S311: removing random noise and environmental noise contained in the enhanced grating signal, the process is as follows: The collected The enhanced grating signals are divided into N groups, and the five-point cubic polynomial center moving smoothing method is used on the N groups to remove the random noise of the enhanced grating signals, which is expressed as: （11）； In the formula, represents the i-th data point after the five-point cubic polynomial center moving smoothing process; represents the i-th data point in the enhanced grating signal; i=1,2,3,…,5; Step S312: using a linear regression analysis method to study the relationship between the environmental noise factor and the support factor, and removing the environmental noise in the enhanced grating signal; the linear regression analysis method includes a univariate linear regression analysis method and a multivariate linear regression analysis method; The univariate linear regression analysis method is expressed as: （12）； In the formula, represents the dependent variables related to ambient noise affected by temperature load and wind speed load; and All represent regression coefficients; Represents the temperature load and wind speed load noise factors in the environmental noise; represents the random error term; Assumptions , then the random variable ; represents normal distribution; represents the standard deviation of the random error term; The multiple linear regression analysis method is expressed as: （13）； In the formula, represents the regression coefficient; Represents the temperature load noise factor in the ambient noise; Step S313: removing the abnormal part of the enhanced acoustic emission signal caused by external environmental interference, the process is as follows: The enhanced acoustic emission signal obtained contains ; represents the nth enhanced acoustic emission signal; Enhanced AE signal average It is expressed as: （14）； In the formula, represents the i-th enhanced acoustic emission signal; Enhanced acoustic emission signal residual error It is expressed as: （15）； According to the Saibel formula, the standard deviation of the enhanced acoustic emission signal is obtained It is expressed as: （16）； With residual error and standard deviation The ratio is judged, and the ratio is expressed as: （17）； when , which means that the enhanced acoustic emission signal detected is interfered by the external environment, and the abnormal part of the enhanced acoustic emission signal caused by the external environment interference is removed; It is the ratio of the residual error to the standard deviation.

9. A bridge monitoring method according to claim 8, characterized in that: In step S4, the fused signal value is obtained, and the specific process is as follows: The BP neural network is used to fuse the processed enhanced acoustic emission signal and enhanced grating signal to obtain the fused signal value; the specific process is as follows: A three-layer BP neural network is established and trained; the BP neural network consists of an input layer, a hidden layer, and an output layer. As the transfer function, we use As the error back propagation function; where represents the transfer function in the BP neural network; e represents the exponential function; Represents the expected output, and continuously adjusts the weights and thresholds of the BP neural network to minimize the error back propagation function E; Represents the calculation output of the BP neural network; The processed enhanced acoustic emission signal and enhanced grating signal are input into the trained BP neural network for fusion to obtain the fused single environmental factor data; The fused single environmental factor data are fused at the feature level to obtain the fused signal value.

10. A bridge monitoring method according to claim 9, characterized in that: In step S5, the early warning result of the bridge health status is obtained, and the specific process is as follows: Step S51: Compare the signal value after fusion in step S4 with the warning thresholds of the three conditions of "normal, warning, and dangerous", and finally obtain the warning result of the health status of the bridge. The specific process is as follows: Step S511: sorting out the historical data of the strong acoustic emission signal and the enhanced grating signal, removing the anomalies in the historical data, and calculating the change rate of each point in the historical data as a sample; Calculate sample expected value It is expressed as: （18）； In the formula, Indicates the deformation rate of the monitoring point; In formula (11) Select r data as a group, select m groups in total, and get the data ; represents the deformation rate of the i-th monitoring point in the j-th group of data; Indicates the number of data in each data group; Indicates the sequence number of the data group; Indicates the data group selected from the overall data; Calculate the expected value of m groups of data ; represents the deformation rate in the jth group of data, calculated from the mth group of data It is expressed as: （19）； （20）； （21）； （22）； （23）； （24）； （25）； In the formula, represents the average value of m groups of data; Indicates the mth variance of the new data calculated from m groups of data; It represents the sample expected value of m groups of new data after calculating the variance of new data from m groups of data; Indicates the jth variance of the new data calculated from m groups of data; Represents each data point in the jth group of data The difference from the mean value of the group of data The sum of the fourth powers of is divided by r-1; represents sample expectation; Indicates the variance of the new data calculated from m groups of data; It represents the sample variance of the new data after the new data composition variance is calculated from m groups of data; according to and Obtain the entropy value of the cloud model and super entropy It is expressed as; （26）； （27）； According to sample expectations , model entropy and super entropy Obtain the cloud droplet diagram; The bridge safety operation value is set according to the cloud droplet diagram, and the deformation rate of the bridge safety operation value is between; According to the bridge safety operation value, the warning thresholds of the three conditions of "normal, warning, and dangerous" are set. The warning thresholds of the three conditions of "normal, warning, and dangerous" are set to 1.3 times, 1.5 times, and 2 times of the safe state respectively; The real-time monitoring fused signal value is compared with the warning thresholds of the three conditions of "normal, warning, and dangerous" to obtain the warning results of the bridge health status.

Images