A structure impact positioning method based on energy curvature and accumulated error

Abstract

The application relates to the field of impact positioning, and particularly discloses a structure impact positioning method based on energy curvature and cumulative error; the structure is divided into a square grid, and an accelerometer is arranged at the intersection of the grid; when an impact event occurs to the structure, the vibration signals collected by the accelerometer are subjected to zero-mean and normalization pretreatment; the energy of the pretreated vibration signals is calculated, and the time point of sudden increase of the energy is taken as the arrival time point of the vibration signals; according to the arrival time point of the vibration signals and the distance between the impact source and the accelerometer, the positioning cumulative error is calculated, and the structure position corresponding to the minimum cumulative error is the positioning result. Through the application, the impact time point can be obtained while the impact source position is positioned, and the application has the advantages of high positioning precision and fast calculation speed.

Description

Technical Field

[0001] This invention belongs to the field of impact positioning technology, specifically relating to a structural impact positioning method based on energy curvature and cumulative error. Background Technology

[0002] Ships, aircraft, and other engineering structures operate in harsh environments for extended periods and are subjected to the interaction of various loads. Sudden malfunctions (such as loosening of structural components) or low-speed collisions with foreign objects can cause damage to these structures, posing a significant threat to their reliability. Utilizing various acoustic and vibration sensors to locate these impact sources can greatly reduce manpower and economic costs, effectively improving the quality maintenance level of ships, and is of great significance.

[0003] Common impact location algorithms include triangulation, which uses the sensor position as the focal point and calculates the focal length by estimating the time difference between wave arrivals at different sensors, thus plotting two hyperbolas and taking the intersection as the impact source location. The location determined by this method is related to factors such as the accuracy of the time difference estimation, wave propagation speed, and target location, with the accuracy of the time difference estimation having a significant impact. Therefore, accurately extracting the arrival time is crucial for solving the location problem. Researchers often consider a signal amplitude exceeding a certain threshold as the signal arrival time, but this method is not ideal under real-world low signal-to-noise ratio conditions. Therefore, there is an urgent need for an easy-to-operate, real-time, and highly accurate impact location method. Summary of the Invention

[0004] The purpose of this invention is to provide a structural impact positioning method based on energy curvature and cumulative error, so as to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] A structural impact localization method based on energy curvature and cumulative error specifically includes the following steps:

[0007] S1. Signal Acquisition: Divide the structure into a square grid, select the intersection of the grid to place accelerometers, and mark the coordinates of the accelerometer positions;

[0008] S2. Signal preprocessing: After an impact event occurs in the structure, the vibration signal collected by the accelerometer is preprocessed by zero-mean normalization and normalization.

[0009] S3. Determine the time of arrival (TOA): Calculate the energy of the preprocessed vibration signal. The moment when the energy suddenly increases is taken as the arrival time of the vibration signal.

[0010] S4. Calculate the location of the impact source: Based on the arrival time of the vibration signal and the distance between the impact source and the accelerometer, calculate the cumulative positioning error. The structural position corresponding to the minimum cumulative error is the positioning result.

[0011] S5. Determine the time of the impact event: Based on the minimum cumulative error value obtained from the solution and the location, calculate the specific time when the impact event occurred.

[0012] Specifically, step S1 includes the following:

[0013] ① The accelerometers are arranged in a rectangular shape;

[0014] ② Establish a rectangular coordinate system in terms of structure, and mark the position coordinates of the i-th accelerometer as (x... i ,y i ).

[0015] Specifically, step S2 includes the following:

[0016] I) Use the following formula (7) to perform zero-mean processing on the vibration signal:

[0017]

[0018] Among them, a i (t) represents the vibration signal collected by the i-th accelerometer, f s Let N be the signal sampling rate, and N be a point in time before the impact event occurs.

[0019] II) Normalize the vibration signal using the following formula (8):

[0020]

[0021] Among them, b i (t) is the vibration signal after zero-mean normalization, max|b i | represents the maximum value of its amplitude. The normalized b i The value of (t) is between 0 and 1.

[0022] Specifically, step S3 includes the following:

[0023] 1) Calculate the energy function E of the vibration signal from the i-th accelerometer using formula (9). i,0 (t);

[0024]

[0025] Among them, s i (τ) represents the preprocessed vibration signal;

[0026] 2) Regarding the energy function E i,0(t) Solving for the first derivative twice consecutively yields the energy curvature function Q. i,1 (t); then, according to formula (10), the energy function E of curvature can be calculated. i,1 (t);

[0027]

[0028] 3) Repeat step 2) of step S3 above four times. According to E i,1 (t) Solve the first derivative twice to obtain Q. i,2 (t), and then according to formula (10), E can be obtained. i,2 (t); according to E i,2 (t) Solve the first derivative twice to obtain Q. i,3 (t), and then according to formula (10), E can be obtained. i,3 (t); according to E i,3 (t) Solve the first derivative twice to obtain Q. i,4 (t), and then according to formula (10), E can be obtained. i,4 (t).

[0029] 4) Take E i,4 (t) The time corresponding to the first non-zero value is taken as the arrival time (TOA) of the sensor signal.

[0030] Specifically, step S4 includes the following:

[0031] a) Based on the arrival time TOA of the signal acquired by the i-th accelerometer i Determine the minimum value among them as TOA min ;

[0032] b) The distance between any point in the structure and its nearest accelerometer is L. r All L r The maximum value in is L max The time t1 of the impact event occurs within the interval Within, where V is the wave propagation speed in the structure; for The result is rounded down and denoted as K;

[0033] c) Within the time interval Within this context, the cumulative error value corresponding to any grid point (x, y) at time k is defined as formula (11):

[0034]

[0035] Where M is the number of accelerometers selected, and (x,y) are the position coordinates of the intersection points of the traversed structure grid;

[0036] d) Compare the cumulative error ε for all k values. k The coordinates of the minimum value (x, y) are the positioning result.

[0037] Specifically, step S5 includes the following:

[0038] The minimum cumulative error value and the coordinates (x, y) of the positioning grid point are calculated according to formula (11), and the corresponding time k value is obtained. Then the time t1 of the impact event can be determined according to the following formula (12).

[0039]

[0040] Compared with the prior art, the beneficial effects of the present invention are:

[0041] This invention utilizes multiple accelerometers. When a structure is impacted, it generates a shock wave, which is detected by accelerometers located at different positions. Furthermore, the invention calculates the arrival time of sensor signals by identifying the moment of energy surge, and then divides the time interval of potential impact events into multiple steps based on the signal sampling rate. Finally, it calculates the cumulative error value. When the cumulative error value is minimized, the most accurate impact location and the most likely impact time can be obtained. Users can accurately determine the impact location, prompting them to conduct more detailed inspections and replace relevant components. This improves structural reliability while reducing maintenance time and costs. Attached Figure Description

[0042] In the picture:

[0043] Figure 1 This is a flowchart of the method of the present invention;

[0044] Figure 2 This is a schematic diagram showing the distance from the impact point to the accelerometer.

[0045] Figure 3 This is a schematic diagram of accelerometer signal acquisition.

[0046] Figure 4 This is a schematic diagram of the signal energy curvature after four cycles.

[0047] Figure 5 This is a schematic diagram showing the signal arrival time;

[0048] Figure 6 A diagram illustrating the division of the impact time period;

[0049] Figure 7 This is a schematic diagram illustrating the minimum cumulative error.

[0050] Figure 8 This is a schematic diagram of the impact positioning results for a flat plate structure. Detailed Implementation

[0051] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0052] Example 1: The structure in this example uses a square steel plate with dimensions of 2m × 2m × 0.02m as a reference, as detailed below:

[0053] Please see Figure 1 - Figure 8 As shown, a structural impact localization method based on energy curvature and cumulative error specifically includes the following steps:

[0054] S1. Signal Acquisition: The steel plate surface is gridded into 64 small grids, and multiple accelerometers are arranged in a rectangular pattern on the structural surface, named A1, A2...A12 respectively. The distance between the accelerometers and the impact point is set to L. I-An ;

[0055] S2. Signal preprocessing: After an impact event occurs in the plate structure, the vibration signal collected by the accelerometer is preprocessed by zero-mean normalization and normalization.

[0056] S3. Determine the time of arrival (TOA), calculate the energy of the preprocessed vibration signal, and take the moment when the energy suddenly increases after the impact as the time of arrival (TOA) of the vibration signal.

[0057] S4. Calculate the positioning position. Based on the arrival time of the vibration signal and the distance between the impact source and the accelerometer, construct a cumulative error function. The structural position corresponding to the minimum cumulative error value is the positioning result.

[0058] Step S1 specifically includes the following steps:

[0059] Reference Figure 2 - Figure 3 When a hammer is used to strike the center of grid number 30, the accelerometers obtain the stress wave signal generated by the impact. Taking accelerometers A1, A3, and A4 located at a distance L as an example, the data sampling rate f... s =25600, the signals collected by accelerometers A1, A3, and A4 are as follows Figure 3 As shown.

[0060] Step S2 specifically includes the following steps:

[0061] Reference Figure 3The signals from sensors A1, A3, and A4 are displayed within a single image. Due to wave attenuation, the signal amplitude decreases with increasing distance. Therefore, A3 captures the signal with the highest amplitude, while A1 captures the signal with the lowest amplitude. The initial amplitude of the captured signal is not exactly zero. Although such a small difference will not affect the accuracy and efficiency of positioning, to avoid potential errors, the signal mean is shifted to zero during signal preprocessing. This is done by subtracting the average acceleration value from the previous 100 moments from the signal acceleration value. After zero-mean preprocessing, the initial amplitude of the signal is close to zero. Simultaneously, normalization preprocessing is performed on the signal; dividing the zero-mean signal by the maximum signal amplitude results in a normalized signal amplitude between 0 and 1.

[0062] The S3 step specifically consists of the following steps:

[0063] 1) The cumulative sum of squares of the acceleration signals A1, A3, and A4 is taken as the signal energy. Each accelerometer experiences a sudden increase in energy upon receiving a bending wave. Therefore, the signal arrival time can be estimated using this abrupt change in signal energy. The rate of change of the energy slope, i.e., the signal energy curvature, is obtained by applying the first derivative to the signal energy twice consecutively.

[0064] 2)Reference Figure 4 To accurately calculate the arrival time of the wave, the cumulative sum of squares of the signal energy curvature is calculated, and then its curvature is determined. Repeating this process four times—first calculating the cumulative sum of squares, then calculating the curvature—reveales that the abrupt change point starts from 0.

[0065] 3)Reference Figure 5 The time of arrival (TOA) of a signal is defined as the time when the signal first becomes non-zero.

[0066] 4) Using the above method, the arrival time (TOA) of all sensor signals can be obtained.

[0067] Step S4 can be further divided into the following steps:

[0068] 1) Divide the detection area into Q uniformly distributed grid points G. i Grid point G i Distance L from different accelerometers Gi-Ai Taking A1 and A2 as examples, the relationship between the accelerometer signal and time can be established to obtain the following error function formula (13):

[0069]

[0070] 2)Reference Figure 6 Since the time t1 of the impact is unknown, for this problem, the impact time t1 falls within a certain time interval [t...]. b TOAmin Therefore, [t] can be placed inside. b TOA min The time interval is divided into several time periods based on the sampling rate. Therefore, for the k-th time period, t1 within this time interval can be represented by the following formula (14):

[0071]

[0072] Therefore, for any time k, the time t1 of the impact event, using sensor combinations A1 and A2, and grid G... i The cumulative error value can be expressed by the following formula (15):

[0073]

[0074] In formula (15), the sampling rate f s =25600, (x1,y1) represents the coordinates of sensor A1, (x2,y2) represents the coordinates of sensor A2, (x Gi ,y Gi ) represents the coordinates of grid point Gi.

[0075] exist Figure 6 In this context, TOA1 is the moment when the wave receives the signal, t1 is the time when the impact event occurs, and TOF is the time required for the wave to travel from the impact point to the accelerometer A1.

[0076] To determine the time interval within which the shock event t1 occurs, we take t b =TOA min -c, where c is the wave propagation time. The distance L between each grid point and its nearest neighbor accelerometer. r In the middle, point P, located at the corner of the structure, is at the maximum distance from sensor A1, with a maximum value of L. max =0.72m. If the curved wave propagates at a speed of V = 500m / s (the wave speed considered is sufficiently conservative), then the time it takes for the curved wave to travel from point P to accelerometer A1 is...

[0077] The minimum TOA (i.e., TOA) min Substituting (=0.4705s) and the approximate wave propagation time into the equation, we can obtain t. b =0.4705 - 0.002 = 0.4685s, indicating that the impact occurred within the time interval [0.4685s, 0.4705s]. However, to increase the reliability of the impact location method, c = 0.01s was chosen for impact location of the plate structure, meaning the impact time t1 occurred within the time interval [0.4605s, 0.4705s]. Based on the data sampling rate f... s=25600, the time interval length is 0.01s. Multiplying the sampling rate and the time interval length gives the number of time points. Therefore, the time interval can be divided into 256 time points, i.e., K=256.

[0078] 4) For all sensor combinations, the cumulative error value for each grid can be expressed by the following formula (16):

[0079]

[0080] In formula (16), if there are M accelerometers, then formula (16) has a total of Each phase summation; according to formula (16), the error of each grid point can be accumulated based on each k value, thereby obtaining the minimum accumulated error value, and thus obtaining the most accurate impact position.

[0081] Reference Figures 7-8 Different values ​​of k will result in different grid points corresponding to the minimum cumulative error. When k represents the actual impact moment, the grid point G... i This represents the actual impact location. Therefore, the cumulative error of the grid points is minimized when the result of the equation is minimized. According to calculations, the minimum cumulative error can be obtained when k=3, so k=3 represents the time when the impact is most likely to occur, thus yielding the most accurate impact location.

[0082] When k=3, the impact time t1=0.4704s, from which the minimum cumulative error of the grid points at this time can be calculated to be 0.0026, which is very consistent with the actual impact location.

[0083] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A structural impact positioning method based on energy curvature and cumulative error, characterized in that, Specifically, the following steps are included: S1. Divide the structure into a square grid, select the intersection of the grid to place the accelerometers, and mark the coordinates of the accelerometer positions; S2. After an impact event occurs in the structure, the vibration signal collected by the accelerometer is preprocessed by zero-mean normalization and normalization. S3. Calculate the energy of the preprocessed vibration signal. The moment when the energy suddenly increases is taken as the arrival time of the vibration signal. The moment when the energy suddenly increases is determined by the following process: 1) Calculate the energy function E of the i-th accelerometer vibration signal using formula (3). i,0 (t); （3） Among them, s i (τ) represents the preprocessed vibration signal; 2) Regarding the energy function E i,0 (t) Solving for the first derivative twice consecutively yields the energy curvature function Q. i,1 (t); then, according to formula (4), the energy function E of curvature can be calculated. i,1 (t); （4） 3) Repeat step 2) above four times, that is: for E i,1 (t) Solve the first derivative twice to obtain Q. i,2 (t), and then according to formula (4), E can be obtained. i,2 (t); for E i,2 (t) Solve the first derivative twice to obtain Q. i,3 (t), and then according to formula (4), E can be obtained. i,3 (t); for E i,3 (t) Solve the first derivative twice to obtain Q. i,4 (t), and then according to formula (4), E can be obtained. i,4 (t); 4) Take E i,4 (t) The moment corresponding to the first non-zero value is taken as the arrival time TOA of the sensor signal; S4. Based on the arrival time of the vibration signal and the distance between the impact source and the accelerometer, calculate the cumulative positioning error. The structural position corresponding to the minimum cumulative error is the positioning result. The method for solving for the minimum cumulative error is as follows: a) The arrival time of the signal acquired by the i-th accelerometer is TOA. i All TOA i The minimum value in is TOA min ; b) The distance between any point in the structure and its nearest accelerometer is Lr, and the maximum value of all Lr is L. max The time t1 of the impact event occurs within the interval Within, where V is the wave propagation speed in the structure; for Round the result and denote it as K, f s The signal sampling rate; c) Within the time interval Within this context, the cumulative error value corresponding to any grid point (x, y) at time k is defined as formula (5): (5) Where M is the number of accelerometers selected, and (x,y) are the position coordinates of the intersection points of the traversed structure grid; d) Compare the cumulative error ε for all k values. k The coordinates of the minimum value (x, y) are the location result. S5. Based on the calculated minimum cumulative error value and the location coordinates, determine the time of the impact event. The method for determining the impact event occurrence time t1 is as follows: According to formula (5) in step S4, by calculating the value of the minimum cumulative error, the coordinates (x, y) of the positioning grid point can be obtained, thus obtaining the time k corresponding to formula (5). Then, the time t1 of the impact event can be determined according to formula (6): 。 2. The method as described in claim 1, characterized in that: The positioning method is applicable to two-dimensional planar structures; three-dimensional structures need to be transformed into two-dimensional structures.

3. The method as described in claim 1, characterized in that: In step S1, a rectangular coordinate system is established structurally, with the accelerometers arranged in a rectangle. The position coordinates of the i-th accelerometer are denoted as (x...). i ,y i ).

4. The method as described in claim 1, characterized in that: In step S2, the vibration signal is zero-mean processed using formula (1): （1） Among them, a i (t) represents the vibration signal collected by the i-th accelerometer, f s Let N be the signal sampling rate, and N be a point in time before the impact event occurs.

5. The method as described in claim 4, characterized in that: In step S2, the vibration signal is normalized using formula (2): （2） Among them, b i (t) is the vibration signal after zero-mean normalization, max|b i | represents the maximum value of its amplitude.

Images