模拟退火算法在地震定位中的应用

阐述了模拟退火算法及其在地震定位中的原理,以震相观测到时与理论到时之差(到时残差)为目标函数,利用模拟退火算法求解目标函数的最小值以获得地震的震源参数。在模拟退火算法的地震定位中根据震相的可靠程度对不同震相设置了相应的定位权重;对震相的走时误差设置了3组不同的值,以分析走时误差对定位结果的影响,对随机生成的1000个模拟地震进行模拟退火算法再定位。结果表明,走时误差越小,定位结果越准确。最后,通过地震震例进一步验证了模拟退火算法在地震定位中的可用性。

The Application of Simulated Annealing Algorithm in Seismic Location

This paper expounds the Simulated Annealing Algorithm(SAA) method and its principle in earthquake location, establishes the difference between observed time and theoretical arrival time (arrival time residual) as the objective function, and uses SAA to solve the minimum value of the objective function in order to obtain the source parameters of earthquakes. In the seismic location based on SAA, the corresponding location weights are set according to the reliability of seismic phases. The accuracy of seismic phase identification determines the accuracy of seismic location to a certain extent. Therefore, three sets of different values are set for the travel-time errors of seismic phases in order to analyze the influence of travel-time errors on the location results, and the SAA is used to relocate 1000 random simulated earthquakes. The results show that the smaller the travel time error is, the more accurate the positioning result is. Finally, the applicability of SAA in earthquake location is further verified by an earthquake example.