基于人工蜂群算法-逐步回归模型的大坝变形监测

Dam Deformation Monitoring Based on Stepwise Regression Model with Artificial Bee Colony Algorithm

在大坝变形监测统计模型研究的基础上,针对传统大坝变形监测回归模型存在的不足,将逐步回归模型与智能优化算法相结合,提出了一种基于人工蜂群算法-逐步回归分析的大坝变形监控模型。该模型以逐步回归方法为基础,利用相关性分析、多重共线性分析等方法对观测数据进行处理,进而对大坝回归模型的荷载集变量进行筛选和评价,并将改进的人工蜂群算法引入回归模型分析,对荷载集系数进行优化和重新评估。人工蜂群算法是一种新型的群体智能优化方法,具有全局智能性搜索、鲁棒性强等优点,将其引入大坝安全监控建模领域,同时为改进人工蜂群算法的局部搜索性能,引入了单纯形操作算子。实例分析表明,与同类模型相比,所提出模型在一定程度上改善了拟合效果,达到了简化模型、提高拟合精度和增强模型预测能力的目的。

Aiming at the deficiency of traditional dam deformation monitoring, a stepwise regres- sion model with artificial bee colony algorithm （ABCA） is proposed for dam deformation monito- ring in this paper. Firstly, the predictor variables are selected by stepwise regression, and the main statistical methods used include： correlation analysis, multiple co-linear relation and so on. Then the coefficients of the regression model are reevaluated by an improved artificial bee colony algorithm. ABCA is a novel swarm intelligence optimization algorithm, and it is introduced into dam health monitoring in this paper. Meanwhile, to improve the performance of local search of ABCA, the simplex operators are introduced into the basic ABCA. Examples show that the pro- posed model for dam deformation monitoring is more effective compared to similar models, con- sidering it can simply the prediction model and improve the accuracy of fitting and prediction.