论文部分内容阅读
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。
Due to the complicated influence factors, there are few researches on long-term settlement prediction model of tunnels. In order to solve the problem of Dunhill prediction of settlement model parameters by nonlinear regression method, Intelligent Single Particle Optimization (ISPO) algorithm was introduced based on Dunwell model, which was applied to a tunnel in Singapore, Shanghai Yan’an East Road Tunnel F19, Shanghai Metro No.1 Long-term settlement prediction of line N12. The results show that the standard deviation of ISPO predictive value and non-linear regression predictive value is reduced by nearly 1 time, which not only overcomes the difficulty of mathematical model parameters but also overcomes the difficult establishment of objective function, The long-term post-construction settlement provides a new way of thinking.