摘要
野马优化算法(Wild Horse Optimizer,WHO)是一种新颖的智能优化算法,瑞雷面波频散曲线反演则是一个多参数、多极值的复杂迭代优化问题.本文将WHO算法引入瑞雷面波频散曲线反演计算,提出改进措施来提高算法局部搜索能力,重点采用反演成功率评价算法在瑞雷面波频散曲线反演中的表现.在三层理论模型试算中,WHO算法的反演成功率仅为52%、43%、37%,而改进野马优化算法(Modified Wild Horse Optimizer,MWHO)的反演成功率分别提升到62%、80%、63%.对比粒子群算法(Particle Swarm Optimizer,PSO)利用四层理论模型进行了试算,PSO算法的反演成功率为53%、63%、59%,MWHO算法的反演成功率均相对较高,分别为60%、75%、63%.在抗噪能力对比中,MWHO算法的表现优于WHO和PSO算法,MWHO算法反演成功率为67%,PSO算法反演成功率为53%,WHO算法反演成功率为42%.在此基础上,对实测瑞雷波数据进行反演计算,MWHO算法的表现依然优于WHO和PSO算法,MWHO算法反演成功率为55%,PSO算法反演成功率为46%,WHO算法反演成功率为41%.理论模型和实测资料的试算表明:MWHO算法在反演成功率、计算精度等方面存在优势,具有一定的实用与研究价值.
WHO(Wild Horse Optimizer)is a novel intelligent optimization algorithm.The Rayleigh wave dispersion curve inversion is a complex iterative optimization problem with multi-parameter and multi-pole.WHO was introduced into the Rayleigh wave dispersion curve inversion in this study,where a modification was presented to improve the local search ability of the algorithm,and the inversion success rate was also proposed to evaluate the performance of the algorithm in the inversion of Rayleigh wave dispersion curve.Among the calculation of three-layer theoretical models,the inversion success rates of WHO were just 52%,43%and 37%.However,the inversion success rates of MWHO(Modified Wild Horse Optimizer)were increased to 62%,80%and 63%.In the calculation of four-layer theoretical models,the inversion success rates of PSO(Particle swarm optimizer)were 53%,63%and 59%,MWHO were raised to 60%,75%and 63%.In contrast to the calculation of theoretical model with noise,the success rate of MWHO was higher than that of WHO and PSO:the success rate of MWHO was 67%,the success rate of PSO was 53%,and the success rate of WHO was 42%.On this basis,WHO,PSO and MWHO were used to calculate the measured Rayleigh wave data.The success rate of MWHO was still higher than that of WHO and PSO:the success rate of MWHO was 55%,the success rate of PSO was 46%,and the success rate of WHO was 41%.The trial calculation statement of theoretical models and measured data:MWHO has advantages in inversion success rate,calculation accuracy and other aspects,and also has certain practical and research value.
作者
唐塑
武银婷
TANG Su;WU YinTing(Urumqi Natural Resources Comprehensive Investigation Center,China Geological Survey,Ürümqi 830057,China;College of Geological Engineering and Geomatics,Chang'an University,Xi'an 710061,China)
出处
《地球物理学进展》
CSCD
北大核心
2024年第4期1698-1710,共13页
Progress in Geophysics
基金
中国地质调查局项目(DD20230380,DD20230060,DD20208081)
陕西省自然科学基础研究计划项目(2022JM-139)资助。
关键词
瑞雷面波
频散曲线
反演
成功率
改进
野马优化算法
Rayleigh wave
Dispersion curve
Inversion
Success rate
Modification
Wild horse optimizer