基于遗传算法和有限混合分布的应力谱多模态建模

发布时间：2019-08-10 08:44

【摘要】：该文提出了一种基于遗传算法(genetic algorithm,GA)的有限混合分布参数估计方法,应用该方法对青马大桥典型焊接节点的应力谱进行多模态建模。首先,采用小波变换消除原始应变监测数据中的温度影响,利用雨流计数法将应变时程曲线转化为日应力谱,考虑到交通荷载(包括汽车荷载和火车荷载)和台风的影响,建立标准日应力谱。然后,采用三种不同的有限混合分布函数(有限混合正态分布函数、有限混合对数正态分布函数和有限混合威布尔分布函数)以及基于遗传算法的混合参数估计方法对应力幅进行多模态建模,根据赤池信息准则(Akaike’s information criterion,AIC)确定最佳的有限混合模型。最后,采用双变量有限混合分布和基于遗传算法的混合参数估计方法建立了应力幅和平均应力二维随机变量联合概率密度函数。结果表明,该文提出的基于遗传算法的有限混合分布参数估计方法可以有效应用于二维随机变量的概率建模。
[Abstract]:In this paper, a finite hybrid distribution parameter estimation method based on genetic algorithm (genetic algorithm,GA) is proposed, which is used to model the stress spectrum of typical welded joints of Qingma Bridge. Firstly, wavelet transform is used to eliminate the temperature influence in the original strain monitoring data, and the strain time history curve is transformed into daily stress spectrum by rain flow counting method. Considering the influence of traffic load (including vehicle load and train load) and typhoon, the standard daily stress spectrum is established. Then, three different finite mixed distribution functions (finite mixed normal distribution function, finite mixed lognormal distribution function and finite mixed Weibull distribution function) and hybrid parameter estimation method based on genetic algorithm are used to model the force amplitude, and the optimal finite mixed model is determined according to the Chichi information criterion (Akaike's information criterion,AIC. Finally, the joint probability density function of two-dimensional random variables of stress amplitude and average stress is established by using bivariable finite mixed distribution and hybrid parameter estimation method based on genetic algorithm. The results show that the finite hybrid distribution parameter estimation method based on genetic algorithm can be effectively applied to the probability modeling of two-dimensional random variables.
【作者单位】：香港理工大学深圳研究院;福州大学土木工程学院;浙江大学建筑工程学院;香港特别行政区政府路政署;
【基金】：国家自然科学基金项目(51308493) 教育部高等学校博士学科点专项科研基金课题项目(20130101120080) 中央高校基本科研业务费专项资金项目(2013QNA4023)
【分类号】：TP18;U441

【相似文献】