地下水水位时间序列中的混沌特征

Chaotic analysis of groundwater time series

利用相空间重构技术、G -P算法以及Wolf提出的从单变量中提取Lyapunov指数的方法 ,分别计算了太原盆地 4组地下水水位时间序列的关联维数和Lyapunov指数。结果揭示在自然状态下 ,太原盆地某些孔隙潜水和孔隙承压水水位变化存在明显的混沌特征。这不仅为以后建立水位预测模型提供了理论依据 ,并且也为运用分形理论研究地质参数的时空变异性以及含水层结构的自相似性提供了进一步的支持。

Firstly, power spectrum of periodic, random, chaotic time series were analyzed. Secondly,correlation dimension and Lyapunov exponent were computed by the reconstruction of phase-space, G-P algorithm and method of determining Lyapunov exponent from time series for identifing chaotic character of groundwater level in Taiyuan basin. Results showed that the chaotic character of groundwater level may exist under the natural condition, and chaotic character of groundwater level series not only can offer theoretical basis for prediction model, but also can provide farther supports for application of fractal theory in spatial and temporal variability of geological parameters and self-resemblance of aquifer constructure.