基于聚类算法的拱坝热学参数反演分析

Inversion Analysis of Thermal Parameters of Arch Dam Based on Clustering Algorithm

由于拱坝坝身较薄,在运行期内坝体温度易受水位、水温和气温等因素的影响,导致坝体温度荷载难以精确掌控,而温度荷载又是拱坝所受到的主要荷载之一。为此,基于实测数据,采用聚类分析与热学参数反演的方法重构整体温度场。首先利用K-Shape聚类算法对坝体上下游方向温度测点进行分类,再依据测点位置实现上下游坝面分区,按照分区情况设置拱坝运行期的上下游温度边界条件;接着引入鲸鱼寻优算法,基于多测点温度实测值,反演坝体热学参数得到坝体整体温度场。工程实例分析结果表明,聚类分析与鲸鱼优化算法的综合应用提升了热学参数反演的效率与精度,反演的参数符合大坝的实际情况。

Due to the thin body of arch dam,the dam temperature is susceptible to water level,water temperature and air temperature during the operation period,so the dam temperature load is difficult to control accurately,but the temperature load is one of the main loads of arch dam.To this end,the overall temperature field of arch dam is reconstructed by clustering analysis and inversion of thermal parameters based on the measured data.Firstly,the K-Shape clustering algorithm is used to classify the temperature test points in the upstream and downstream directions,and then the upstream and downstream dam surface partition is realized,and the upstream and downstream temperature boundary conditions of the arch dam during operation period are set according to the partition conditions.Then,the whale optimization algorithm is introduced to obtain the overall temperature field by the inversion of thermal parameters based on the measured temperature value of multiple test points.The engineering example analysis results show that the comprehensive application of the cluster analysis and the whale optimization algorithm can improve the efficiency and accuracy of thermal parameter inversion,and the inversion results meet the actual situation of the dam.