基于FOA-BP-AdaBoost的大坝变形预测模型及应用
Dam Deformation Prediction Model Based on FOA-BP-AdaBoost and Its Application
王凯 1李鸳承 2范亚军 1何广焕 1蒙金龙 1赵磊1
作者信息
- 1. 广西建设职业技术学院,广西 南宁 530007
- 2. 广西职业师范学院,广西 南宁 530007
- 折叠
摘要
为提升大坝变形监测预测精度,解决变形量受多因素影响等问题,笔者提出了基于果蝇优化算法(FOA)、BP神经网络的AdaBoost强预测组合模型(FOA-BP-AdaBoost),并与BP神经网络模型、FOA-BP神经网络模型应用于工程实例中的预测精度进行多方位量化对比.结果表明:强预测模型集齐了果蝇算法全局优化、BP神经网络局部寻优和AdaBoost"优中选优"的特点,最大程度优化了预测效果;实例应用证实了FOA-BP-AdaBoost模型在大坝变形预测领域的准确性和有效性.该模型已成功应用于工程实例,可为类似工程提供参考.
Abstract
In order to improve the prediction accuracy of dam deformation monitoring and solve the problem that the deformation is affected by many factors,the AdaBoost strong prediction combination model based on fruit fly optimization algorithm(FOA)and BP neural network(FOA-BP-AdaBoost)is proposed,and compared with the prediction accuracy of BP neural network model and FOA-BP neural network model applied to engineering examples.The results show that the strong prediction model integrates the characteristics of global optimization of fruit fly algorithm,local optimization of BP neural network and AdaBoost"optimal selection",and optimizes the prediction effect to the greatest extent;The application of the example confirms the accuracy and effectiveness of the FOA-BP-AdaBoost model in the field of dam deformation prediction.The model has been successfully applied to engineering examples,which can provide reference for similar projects.
关键词
大坝/变形监测/FOA-BP-AdaBoost模型/强预测模型/果蝇优化算法/BP神经网络Key words
dam/deformation monitoring/FOA-BP-AdaBoost model/strong prediction model/fruit fly optimization algorithm/BP neural network引用本文复制引用
基金项目
广西壮族自治区高等学校中青年教师科研基础能力提升项目(2020KY35022)
出版年
2024
国家科技期刊平台
NSTL
万方数据