基于差分进化算法的微震定位

Microseismic positioning method based on differential evolution algorithm

  • 摘要: 针对传统基于走时微震定位方法存在的定位精度依赖于走时精度问题,提出了改进定位目标函数的差分进化微震定位方法。基于传统微震P波走时定位方法,在定位目标函数中加入了误差加权系数,通过降低初至拾取精度低的微震信号道对定位误差的影响,减弱了初至拾取精度对微震定位的影响;利用差分进化算法,通过种群初始化、变异、交叉和选择等操作,求解改进的微震定位目标函数,实现微震定位;对模型数据和实测数据进行试处理,结果表明:改进的目标函数能明显降低误差较大分量对总误差的贡献;提出的改进定位目标函数的差分进化微震定位方法,在部分微震信号道走时拾取误差较大甚至错误的情况下,能很大程度减弱拾取误差对定位精度的影响,可以获得较好的定位结果。

     

    Abstract: Aiming at the problem that the location accuracy of conventional time-based microseismic location method depends on the accuracy of travel time, a differential evolution microseismic location method is proposed to improve the locating target function. Based on the traditional micro-seismic P-wave travel-time location method, the error weighting coefficient was added to the localization objective function, which reduced the impact of the first-time picking accuracy on the microseismic location by reducing the effect of the microseismic signal path with low initial pick-up accuracy on the locating error. The differential evolution algorithm, through population initialization, mutation, crossover, and selection operations, solves the improved micro-seismic location target function and achieves microseismic location. The model data and measured data are subjected to trial processing. The results show that the improved objective function can significantly reduce the error. The contribution of large components to the total error; the proposed differential evolution micro-seismic location method to improve the localization objective function can greatly reduce the influence of the pick-up error on the locating accuracy when the pick-up error of some microseismic signal paths is large or even wrong, get better location results.

     

/

返回文章
返回