A two-dimensional sensor network layout method for optimizing acoustic emission positioning accuracy

Abstract

The application discloses a two-dimensional sensor network layout method for optimizing acoustic emission positioning precision, establishes a seismic source analytical positioning model under non-velocity measurement conditions, combines a plurality of sensor layout comparison experiments, and reveals a cooperative optimization mechanism of a sensor envelope range and a regional interconnection density on positioning precision. Experiments show that the sensor combination numbered 2, 3, 5, 6 and 9 can realize high-precision positioning with an average error of less than or equal to 2.5 mm, and provides a reliable technical scheme for rock mass engineering fracture monitoring.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of non-destructive testing of rock materials, and particularly relates to a two-dimensional acoustic emission source positioning method based on sensor network layout optimization under non-velocity measurement conditions, which is suitable for high-precision positioning of a fracture source in rock mass engineering. BACKGROUND

[0002] Rock fracture instability is an important inducement of rock mass engineering disasters (such as rock burst and slope instability), and accurate monitoring of the internal crack evolution mechanism of the rock mass is crucial for disaster prevention and control. The acoustic emission technology can realize dynamic tracking of the fracture source position by capturing the elastic wave signals generated by rock fracture. However, the existing acoustic emission positioning methods are mostly based on a single wave velocity model, and the anisotropy of rock materials leads to significant differences in wave velocity propagation paths, resulting in large positioning errors of the traditional methods.

[0003] In the prior art, although the Kundu method, the Ing l ada method and the USBM method treat the wave velocity as an unknown quantity to improve the accuracy, they still rely on iterative algorithms, which have high computational complexity and are easily affected by the initial value. In addition, the influence of sensor layout on positioning accuracy lacks systematic research, especially under non-velocity measurement conditions, the correlation mechanism between the spatial distribution of the sensor network and the positioning accuracy has not been clearly defined. SUMMARY

[0004] The application proposes a two-dimensional positioning method based on sensor network layout optimization to solve the problem of insufficient acoustic emission positioning accuracy under non-velocity measurement conditions in the prior art. The optimal sensor configuration scheme is determined by combining theoretical algorithms with experimental verification.

[0005] TECHNICAL SCHEME

[0006] The core of the application is to establish an analytical model for source positioning under non-velocity measurement conditions, and to reveal the synergistic influence law of the sensor envelope range and the regional connection density on the positioning accuracy by comparing multiple sensor layout experiments, and finally determine the high-precision layout mode. The specific steps are as follows:

[0007] 1. Positioning algorithm construction

[0008] Based on the unknown parameters of the source position (x, y) in the two-dimensional plane, the time node t0 generated by the source and the wave velocity v, the sensor distance equation is established:

[0009]

[0010] Through linear equation conversion and Cramer's rule, a unique analytical solution is obtained, and the minimum number of sensors is 5.

[0011] 2. Experimental device and process

[0012] Device: 520mmx400mmx5mm marble plate, 6 ISAS03-150 piezoelectric sensors, 40dB preamplifier, AMSY-6 acoustic emission analyzer.

[0013] Flow:

[0014] a. Establish a coordinate system with the center of the marble plate as the origin, simulate the seismic source with a broken lead (diameter 0.5mm, HB hardness), and preset the coordinates of the broken lead point;

[0015] b. 6 sensors are arranged according to the preset layout, and the signals collected are amplified and recorded by the AMSY-6 acoustic emission analyzer;

[0016] C. Extract the arrival time of each sensor signal, select 5 kinds of sensor combinations (number 1, 2, 3, 4, 5, etc.), and calculate the source coordinates and wave velocity;

[0017] d. Eliminate invalid data with wave velocity exceeding 3700-4300m / s, calculate the positioning error and standard deviation, and draw a heat map to analyze the effective positioning range.

[0018] 3. Layout optimization criteria

[0019] Sensor envelope range: the larger the coverage area, the higher the positioning accuracy;

[0020] Area internal wiring density: the denser the distribution of sensors near the diagonal line in the target area, the more significant the accuracy improvement.

[0021] Beneficial effects

[0022] The present application optimizes the sensor network layout and constructs an analytical positioning model under non-velocity measurement conditions, significantly improving the accuracy and reliability of acoustic emission source positioning, which is embodied in the following aspects:

[0023] ① High-precision positioning. Through the coordinated optimization of sensor envelope range (≥80% coverage) and area internal wiring density (≥3 sensors in key areas), the experimental verification shows that the average positioning error of the optimal layout combination is reduced to 6.23mm, and the standard deviation is 3.8mm, which is more than 40% higher than the accuracy of the traditional method;

[0024] ② Simplify the operation process. Without the need for pre-measurement of wave velocity, the source coordinates and wave velocity are directly solved by the analytical model, eliminating the wave velocity error caused by the anisotropy of the rock, reducing the artificial intervention and calculation complexity;

[0025] ③ Wide applicability. It is suitable for anisotropic materials such as rocks, and can effectively monitor the rupture source in engineering disasters such as rock burst and slope instability, providing reliable technical support for disaster warning;

[0026] ④High efficiency and economy. The analytical model is based on the Kramers rule to achieve fast solution, avoiding the time-consuming problem of iterative algorithm, and at the same time, only 5 sensors are needed to complete positioning, reducing the cost of equipment deployment;

[0027] ⑤Data-driven optimization. Through error screening (removing data with wave velocity outside the range of 4000±300 m / s) and heat map analysis, the positioning accuracy and effective positioning area are determined, enhancing the pertinence and reliability of engineering application.

[0028] The technical scheme of the present application has both theoretical innovation and engineering practicability, and provides an efficient and low-cost technical means for non-destructive testing and safety monitoring of rock mass engineering. BRIEF DESCRIPTION OF DRAWINGS

[0029] In order to more clearly illustrate the technical scheme of the embodiments of the present application, the drawings of the embodiments will be briefly introduced below. Obviously, the drawings described below only relate to some embodiments of the present application, but not limit the present application.

[0030] Figure 1 : experimental equipment connection schematic diagram;

[0031] Figure 2 : sensor layout and broken lead point coordinate distribution (5 groups of comparison);

[0032] Figure 3 : positioning result scatter plot of 5 groups of experiments;

[0033] Figure 4 : comparison of average positioning error and standard deviation;

[0034] Figure 5 : positioning heat map, reflecting the effective coverage range of different layouts. DETAILED DESCRIPTION

[0035] 1. A marble flat plate sample with a size of 520mmx400mmx5mm is selected, and a rectangular coordinate system is established with the center of the sample as the origin. A lead is broken on the sample to simulate a seismic source, and the diameter of the lead core used for breaking the lead is 0.5mm, and the hardness is HB. When breaking the lead, the lead core is stretched out by 2.5mm and kept at an angle of 30° with the sample surface.

[0036] 2. Six acoustic emission sensors are dispersedly arranged on the surface of the sample to detect acoustic emission source signals, the seismic source signal is amplified by a preamplifier with an amplification of 40dB, transmitted to a vallen acoustic emission signal analyzer (AMSY-6), and the acoustic emission information of the whole process is recorded through a computer. After obtaining the waveform data of the acoustic emission signal, the arrival time of each sensor can be picked up by using the analysis software provided by the system.

[0037] 3. After each lead break, the sensors will receive the acoustic emission signals generated by the lead break. According to the time difference of the signals received by the sensors, 5 sensors with irregular broken line distribution are selected to form a group, and the coordinates of different lead break positions and wave velocity v are calculated. The calculated coordinates are compared with the lead break coordinates to analyze the positioning accuracy of the sensors under different arrangement modes.

[0038] 4. From the 6 sensors laid out, 5 are selected to form a sensor array for source positioning inversion, a total of 5 groups. The first group of array sensor numbers is 1, 2, 3, 4, and 5; the second group of array sensor numbers is 1, 2, 3, 4, and 6; the third group of array sensor numbers is 1, 2, 3, 5, and 6; the fourth group of array sensor numbers is 1, 2, 4, 5, and 6; the fifth group of array sensor numbers is 2, 3, 4, 5, and 6.

[0039] 5. The wave velocity v of the sample is in the range of 4000±300 m / s, so in the calculation process, the data of v<3700 m / s and v>4300 m / s are considered as invalid positioning. In order to ensure the effectiveness of the experimental results, these data are excluded from the positioning results.

[0040] 6. After excluding invalid coordinates, the positioning results of the 5 groups of experiments are obtained( Figure 3 ), the error levels of each group of experiments are compared, and according to the distance error between the positioning point coordinates calculated by each lead break and the lead break position, the average error and standard deviation of each positioning experiment are calculated( Figure 4 ), and the positioning accuracy is analyzed accordingly.

[0041] 7. The heat map can be used to intuitively see the correlation between data. In the positioning experiment, the closer the positioning result is to the source position, the stronger the correlation is. Therefore, the source positioning heat map of each group of experiments is made( Figure 5 ) to determine the effective positioning range of each group of experiments.

[0042] 8. In the experiment, the second group of sensor layout (numbers 1, 2, 3, 4, and 6) has an envelope coverage rate of 85% and the highest target area connection density, with an average positioning error of 6.23 mm (standard deviation 3.8 mm), significantly better than other combinations. The heat map shows that the effective positioning area is expanded by 35%, verifying the reliability of the layout optimization criteria.

Claims

1. A method for optimizing the layout of a two-dimensional sensor network for acoustic emission localization accuracy, characterized in that The method comprises the following steps: Six acoustic emission sensors numbered 1, 2, 3, 4, 5 and 6 are arranged on the marble surface, a rectangular coordinate system is established with the center of the sample as the origin, the sensor numbers and positions are shown in FIG. 2, five sensors are selected from the six arranged sensors to form a sensor array, and a total of five groups are formed, wherein the first group of array sensors are numbered 1, 2, 3, 4 and 5; the second group of array sensors are numbered 1, 2, 3, 4 and 6; the third group of array sensors are numbered 1, 2, 3, 5 and 6; the fourth group of array sensors are numbered 1, 2, 4, 5 and 6; and the fifth group of array sensors are numbered 2, 3, 4, 5 and 6; Based on non-measuring speed condition, an analytical model of seismic source positioning is established. Under the condition that each sensor receives the time node ti of the seismic source signal and the corresponding sensor coordinates Si(xi, yi), the seismic source coordinates (x, y), the time node t0 generated by the seismic source, and the wave speed v are solved by simultaneously solving the sensor distance equation, where i = 1, 2, …, 6, and the sensor distance equation is The analytical model is solved by Cramer's rule for a linear equation set, which is in the form of AX = B, wherein the matrix A and the vector B are constructed by the sensor coordinates and the arrival time difference, and the solution vector X includes x, y, v, and t0. The arrival time of each sensor is collected through the lead breaking experiment, the coordinates of different lead breaking positions are calculated, the obtained coordinates are compared with the lead breaking positions, and the positioning error is calculated; According to the cooperative influence of the sensor envelope range and the line density in the region on the positioning accuracy, the optimal layout is determined as the combination of sensor numbers 1, 2, 3, 4 and 6.

2. The method of claim 1, wherein In the sensor layout, the envelope range covers more than 80% of the area of the marble plate, and each unit area in the target region contains at least two sensors.

3. The method of claim 1, wherein The positioning results obtained by excluding invalid coordinates are compared with the error levels of each group of experiments to analyze the positioning accuracy.

4. The method of claim 1, wherein The correlation between the positioning results and the source positions is analyzed by heat map, effective positioning areas are screened, and data points outside the range of 4000±300 m / s of wave velocity are removed.

Images