Localization of microseismic source based on genetic-simplex hybrid algorithm

• Published in: 2017 Chinese Automation Congress (CAC) pp. 4002 - 4007
• Main Authors: Yijia Li, Qingmei Sui, Jing Wang, Zhengfang Wang, Lei Jia, Hanpeng Wang, Shengsan Shen, Bo Du
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.10.2017
• Subjects: Band-pass filters; Coal mining; Filter; Finite impulse response filters; Genetic-simplex hybrid algorithm; Linear programming; Localization; Microseismic(MS); Monitoring; Sensors
• DOI: 10.1109/CAC.2017.8243480

The microseismic (MS) monitoring in the coal mining provides a powerful method to identify the coal mine geological hazards as well as to prevent against the illegal mining. The challenge of MS monitoring is to extract characteristics of interest from the collected raw MS signals and locate the MS events using effective algorithms. In order to improve the localization accuracy of microseismic sources, a genetic-simplex hybrid localization algorithm has been elaborated, which employs Genetic algorithm to determine the initial values and then utilizes simplex algorithm to search the optimal locations of MS events. Meanwhile, frequency filtering method is employed to extract the specific frequency components from the raw MS signals so as to calculate the time differences. Experiments have been performed on a planar with the dimension of 5m*5m to validate the proposed method. The experimental results indicate that the hybrid algorithm provides higher localization accuracy compared with the traditional simplex algorithm. The maximum radial error of the hybrid localization algorithm is 0.47 m, while it is 0.94m for the traditional simplex algorithm. And the relative position error of the hybrid positioning algorithm is less than 10%, which is around 10% higher than the traditional simplex algorithm.