72
Q. Hu and L. Dong
propose an AE source location method to improve the location accuracy in the real
complex structure, which is different from the conventional methods.
In this study, an Ant Colony Localization Method (ACLM) without premeasured
velocity was proposed, which is based on the ant colony algorithm [10] and the
grid method. The accuracy and applicability of the ACLM are verified by the leadbreaking test.
7.2 Ant Colony Localization Method (ACLM)
The ACLM uses the ant colony algorithm to search the fastest wave path from
the source to the sensor and determines the AE source according to the relationship
between the fastest wave path and the arrivals, which also does not need to premeasure
the velocity of the complex structure. It can be divided into six parts, which are
detailed below.
7.2.1 Determine Region of Localization
The ACLM is proposed as a source location method for complex structures. It is
suggested to determine the region of interest before localization. Though it is possible
for ACLM to locate the source in the complete structure, it would be time-consuming
and would not improve the accuracy of source location. Therefore, the first step of
ACLM prefers to determine the important region of the structure, including the region
where failure or fracture can be predicted.
7.2.2 Construct Grid and Represent Region
The geometry and specific position of the complex region are determined in the location region. Moreover, the size of the unit square grid is determined according to the
sharp of the complex structure and the accuracy requirement of the source location.
Up to a point, the higher the resolution of the grid, the better the location result of
this method. However, the location accuracy would not be obviously improved when
the grid is sufficiently densely divided. A matrix M with the same size as the grid
node is established, and the matrix index position (i, j) corresponding to the grid
node is marked with the digits 0 or 1 to represent the location region. The grid nodes
P i j form a set G N that serves as the starting point to search for the fastest waveform
path among subsequent nodes. Assuming that the propagation velocity of the P wave
in the surrounding non-empty region is a fixed unknown value and is represented by
C. Because the wave velocity changes with time, temperature, and pressure in the
dynamic monitoring, there’s no need to measure the P -wave velocity in the test. For
Q. Hu and L. Dong
propose an AE source location method to improve the location accuracy in the real
complex structure, which is different from the conventional methods.
In this study, an Ant Colony Localization Method (ACLM) without premeasured
velocity was proposed, which is based on the ant colony algorithm [10] and the
grid method. The accuracy and applicability of the ACLM are verified by the leadbreaking test.
7.2 Ant Colony Localization Method (ACLM)
The ACLM uses the ant colony algorithm to search the fastest wave path from
the source to the sensor and determines the AE source according to the relationship
between the fastest wave path and the arrivals, which also does not need to premeasure
the velocity of the complex structure. It can be divided into six parts, which are
detailed below.
7.2.1 Determine Region of Localization
The ACLM is proposed as a source location method for complex structures. It is
suggested to determine the region of interest before localization. Though it is possible
for ACLM to locate the source in the complete structure, it would be time-consuming
and would not improve the accuracy of source location. Therefore, the first step of
ACLM prefers to determine the important region of the structure, including the region
where failure or fracture can be predicted.
7.2.2 Construct Grid and Represent Region
The geometry and specific position of the complex region are determined in the location region. Moreover, the size of the unit square grid is determined according to the
sharp of the complex structure and the accuracy requirement of the source location.
Up to a point, the higher the resolution of the grid, the better the location result of
this method. However, the location accuracy would not be obviously improved when
the grid is sufficiently densely divided. A matrix M with the same size as the grid
node is established, and the matrix index position (i, j) corresponding to the grid
node is marked with the digits 0 or 1 to represent the location region. The grid nodes
P i j form a set G N that serves as the starting point to search for the fastest waveform
path among subsequent nodes. Assuming that the propagation velocity of the P wave
in the surrounding non-empty region is a fixed unknown value and is represented by
C. Because the wave velocity changes with time, temperature, and pressure in the
dynamic monitoring, there’s no need to measure the P -wave velocity in the test. For