摘要
根据大气环境下腐蚀金属的极化行为特征,选择与之相适应的极化曲线方程,方程中包含7个电化学腐蚀动力学参数,如何求得高精度的参数至关重要.极化曲线拟合属于非线性最小二乘问题,而传统的非线性拟合方法来解决曲线拟合问题时有相当明显的缺陷,例如过分依赖参数初始值、拟合精度不高和结果陷入局部最优问题.对此提出基于粒子群-信赖域的极化曲线拟合算法来求得动力学参数,并通过实验证明该方法的有效性和鲁棒性.
According to the polarization behavior of corrosive metals in the atmosphere,the polarization curve equation is selected.The equation contains seven electrochemical corrosion kinetic parameters.It is important to obtain high-precision parameters.Polarization curve fitting belongs to the nonlinear least squares problem,and the traditional nonlinear fitting method has obvious defects when solving the curve fitting problem,such as excessive dependence on the initial value of the parameter,low fitting precision and the result falls into local optimum.A polarization curve fitting algorithm based on particle swarm optimization and trust region was proposed to obtain the dynamic parameters.Experiments show the effectiveness and robustness of the method.
作者
孙峰
孙伟
Sun Feng;Sun Wei(School of Information Engineering,Shanghai Maritime University,Shanghai 201306,China)
出处
《计算机应用与软件》
北大核心
2019年第11期280-285,共6页
Computer Applications and Software
关键词
极化曲线
腐蚀动力学参数
曲线拟合
粒子群算法
信赖域算法
Polarization curve
Corrosion kinetic parameters
Curve fitting
Particle swarm optimization
Trust region algorithm
