摘要
电阻率测量是海底沉积物工程地质勘察的主要原位观测方法之一,作为一种间接测量方法,需要建立沉积物物性参数与沉积物电阻率的回归模型。为提高建模精度,本文提出了一种基于鲸鱼算法优化的最小二乘支持向量机(Whale Optimization Algorithm-Least Squares Support Vector Machine,WOA-LSSVM)的海底沉积物物性参数与电阻率回归建模方法。该方法建立了海底沉积物电阻率与沉积物4种基本物性参数(含水率、密度、孔隙比、塑性指数)的单输入、单输出最小二乘支持向量机(Least Squares Support Vector Machine,LSSVM)回归模型,利用WOA算法对LSSVM参数进行寻优取值。对比研究了WOA算法、遗传算法(Genetic Algorithm,GA)、粒子群(Particle Swarm Optimization,PSO)算法优化的LSSVM建模结果,结果表明,基于WOA-LSSVM建立的海底沉积物物性参数与电阻率的回归模型具有更好的预测效果,均方根误差降低1.1%~14.9%,平均绝对百分比误差降低0.4%~19.9%。
Resistivity measurement is one of the main in-situ observation methods for submarine sediments engineering geological survey. As an indirect measurement method, it is necessary to establish regression models of sediments physical parameters and sediments resistivity. In order to improve the modeling accuracy, this paper proposes a regression modeling method based on the Whale Optimization Algorithm-Least Squares Support Vector Machine(WOA-LSSVM) of submarine sediments physical parameters and resistivity. This method establishes the single input and single output least squares support vector machine(LSSVM) regression models of submarine sediments resistivity and sediments four basic physical parameters(moisture content, density, void ratio,and plasticity index), and uses the WOA algorithm to optimize the LSSVM parameters. The LSSVM modeling results optimized by WOA, Genetic Algorithm(GA) and Particle Swarm Optimization(PSO) are compared and studied. The results show that the regression models of submarine sediments physical parameters and resistivity based on WOA-LSSVM have a better predictive effect,the root mean square error(RMSE) is reduced by 1.1% to 14.9%, and the mean absolute percentage error(MAPE) is reduced by 0.4%to 19.9%.
作者
张帆
刘兰军
陈家林
刘健
ZHANG Fan;LIU Lanjun;CHEN Jialin;LIU Jian(College of Engineering,Ocean University of China,Qingdao 266100,China;Key Laboratory of Marine Environment and Geological Engineering Shandong Province,Qingdao 266100,China;Shandong Provincial Engineering Research Center for Marine Intelligent Equipment Technology,Qingdao 266100,China)
出处
《海洋技术学报》
2021年第4期84-94,共11页
Journal of Ocean Technology
基金
国家重点研究发展计划资助项目(2017YFC0307701)
国家自然科学基金资助项目(41427803)
青岛海洋科学技术国家实验室开放基金资助项目(QNLM2016ORP0414)
