全过程沉降量预测的Logistic生长模型及其适用性研究

Logistic growth model and its applicability for predicting settlement during the whole process

地基的全过程沉降量与时间的关系呈S形曲线,可用Logistic生长模型来描述。目前应用的Logistic生长模型,较少考虑模型的适用范围和预测精度等问题。结合工程实例,比较了模型参数估计的三种方法,研究结果表明,采用非线性回归法估计模型参数,可取得较好的预测效果;根据混沌理论,研究了模型的适用范围和预测性能。Logistic生长模型的动态行为由参数a(瞬时沉降速率)决定,当参数a的取值范围在[0, 2]之内,该模型预测的最终沉降量是稳定的。

The settlement-time relation of foundation during the whole process of construction and operation can be described by a sigmoid or S-shaped curve, which can be modeled by the logistic growth model. However the range of applicability and the prediction accuracy is seldom considered when using this model. According to the field data from practical engineering, Three methods for parameter estimation of the Logistic growth model are compared, and The comparison indicates that the nonlinear regression method has a better accuracy. The applicability and the prediction performance of the model are evaluated by using chaos theory. The dynamic behavior of Logistic growth model is controlled by parameter a, namely transient settlement rate. When parameter a is in the interval [0, 2], the final settlement predicted by Logistic growth model is stable.