论文部分内容阅读
影响地下水位变化因素有很多,在正常情况下,地下水位的变化实际上反应了气压、固体潮和降雨这些因素的变化,但是这些影响因子与地下水位之间有着较强的非线性关系。该文使用支持向量机方法建立起崇明中学观测站地下水位与气压、固体潮和降雨这些因素之间的非线性关系模型,并用于地下水观测数据拟合与预测,得到了较理想的结果,明显优于逐步回归方法。研究结果表明,支持向量机方法在地震前兆数据处理中有着广泛的应用前景。文中还对支持向量机方法在实际应用中的有关问题进行了讨论。
There are many factors influencing the variation of groundwater table. Under normal circumstances, the change of groundwater table actually reflects the changes of air pressure, solid tide and rainfall. However, there is a strong nonlinear relationship between these factors and groundwater table. In this paper, we establish a nonlinear relationship model between groundwater level and atmospheric pressure, solid tide and rainfall in Chongming Middle School Observatory by using SVM method, and apply it to the fitting and prediction of groundwater observation data. The result is obvious Better than the stepwise regression method. The results show that the SVM method has a wide range of applications in seismic precursory data processing. The article also discusses the problems of support vector machine in practical application.