论文部分内容阅读
在地铁工程的设计、施工、工后沉降控制过程中,拱顶下沉监测值是反映地下工程结构安全和稳定的重要数据.针对常用的地铁拱顶沉降测模型只能做短期预测,精度不高,且需要一些土的本构参数的问题,将相空间重构、最小二乘支持向量机理论相耦合,建立基于改进C-C方法相空间重构和最小二乘支持向量机的地铁隧洞拱顶沉降混沌时间序列预测模型.经实例演算,模型比传统C-C方法相空间重构、基于最大Lyapunov指数的混沌预测模型、人工神经网络模型拟合效果好,预测精度高.
During the design, construction and post-construction settlement control of metro project, the monitoring value of vault settlement is an important data reflecting the safety and stability of the underground engineering structure.For the commonly used subway vault settlement model can only make short-term prediction accuracy is not High and needs some constitutive parameters of soil, phase space reconstruction and least square support vector machine theory are combined to establish the subway tunnel vault based on the improved CC method of phase space reconstruction and least square support vector machines Settling chaotic time series prediction model.According to the example calculation, the model is reconstructed phase space compared with the traditional CC method, the chaotic prediction model based on the maximum Lyapunov exponent, artificial neural network model fitting effect is good, and the prediction accuracy is high.