Study And Apply In The Chaotic Time Seriesmethod In The Groundwater Level Prediction

At present, many hydrological forecasting for traditional methods are establishmented of at linear systems, their common drawback is that too much emphasis on the study of mathematics. Chaotic time series forecasting methods as a non-linear forecasting methods, based on chaos theory in order to break through the traditional methods to establish the limitations of the subjective model, through the inherent law of the time series analysis to predict, in many areas has been successfully applied.In this paper, exploring a variety of traditional methods based on chaos theory will be used in hydrological forecasting, through phase-space reconstruction technique, select the appropriate delay time 'Ï„' and embedding dimension 'm', embedded into the ground water level time series phase space of the Chinese Academy of Sciences in Hebei Luancheng Comprehensive Experimental Station The three pilot sites. By calculating the Lyapunov index to determine the positive and negative characteristics of chaotic systems using the weighted zero-rank local-region method and the weighted one-rank local-region method to predict groundwater level in the future. In the numerical realization, the paper design a 'Matlab' software platform C-C scientific method, G -P algorithm, the weighted zero-rank local-region method, and the weighted one-rank local-region method, and so on. Forecast results show that the chaotic time series prediction accuracy and credibility has been enhanced compared traditional forecasting methods. Chaotic time series prediction for the purpose not only to predict, but also to verify the characteristics of chaos theory and remodeling parameters in order to study and improve their algorithms.