摘要
由于影响因素复杂,隧道长期沉降预测模型研究偏少。针对非线性回归法求解邓英尔预测沉降模型参数的不足,在邓英尔模型基础上引入智能单粒子优化算法(ISPO),分别用于新加坡某隧道、上海延安东路隧道F19、上海地铁一号线N12的长期沉降预测。结果表明,ISPO预测值与非线性回归法预测值相比标准差减小了近1倍,既克服了数学模型参数的较难确定又克服了目标函数的较难确立,为模型预测在地铁隧道工后长期沉降中的应用提供了一种全新的思路。
Because of the complex influence factors,prediction models of long-term settlement are relatively less.For that non-linear regression method is insufficient to solve parameters of Deng Yinger settlement prediction model,we introduced the intelligent single particle optimization（ISPO）,based on which predicted long-term settlement of a Singapore tunnel,Shanghai East Yan＇an Road Tunnel F19 and Shanghai subway line 1 N12.Results show that the standard deviation by predictive value of ISOP is nearly half of that by nonlinear regression method.This method overcomes the difficulty to determine mathematical model parameters as well as to establish objective function.Besides,it provides a new way to predict long-term surface settlement after shield tunneling construction.
出处
《武汉理工大学学报》
CAS
CSCD
北大核心
2012年第12期89-92,共4页
Journal of Wuhan University of Technology
基金
国家自然科学基金(51078332
51278463)
浙江省自然科学基金(LQ12E08009
Z1100016)
浙江省教育厅科研项目(Y20122339)
关键词
地铁隧道
长期沉降
邓英尔模型
智能单粒子算法
预测
tunnel
long-term settlement
Deng Yinger model
intelligent single particle optimization（ISPO）
predict
