论文部分内容阅读
强震引发的工程场地地表破裂对其上的建(构)筑物会造成严重破坏.准确预测地震发生后地表破裂带宽度,确定出合理的断层避让距离对工程建设场地的选择具有重要意义.本文提出一种利用信息扩散原理对强震地表破裂宽度进行预测的方法.首先在诸多影响因素分析的基础上选取震级、上覆土层厚度为主要评价指标,采用模糊信息优化处理中的信息扩散原理及一、二级模糊近似推论,建立了两个评价指标与地表破裂宽度之间的模糊关系;然后针对模型的稳定性和精度进行交叉验证,并与多元线性回归模型、完全二次回归模型和BP神经网络模型进行对比,计算结果显示,信息扩散方法构建的强震地表破裂宽度模型可以较好的处理各指标间的非线性关系;最后,以实例分析验证了该方法的有效性和可行性.
The rupture of the construction site caused by strong earthquakes will cause serious damage to the structures on it.It is of great significance to accurately predict the width of the surface rupture zone after the earthquake and to determine the reasonable avoidance distance for the construction site selection. In this paper, we propose a method to predict the surface rupture width of strong earthquakes based on the principle of information diffusion.Firstly, based on the analysis of many influencing factors, we select the magnitude and overburden thickness as the main evaluation indexes, and adopt the principle of information diffusion in fuzzy information optimization And the first and second level fuzzy approximate inference, established the fuzzy relationship between the two evaluation indexes and the width of the surface rupture. Then, the stability and accuracy of the model were cross-validated, and compared with multiple linear regression model, complete quadratic regression model and BP neural network model. The calculated results show that the strong ground surface rupture width model constructed by the information diffusion method can well handle the nonlinear relationship among various indexes. Finally, the effectiveness and feasibility of the method are verified by an example analysis .