Application Of Nonlinear Fractal Theory And Time Series Analysis Method In Prediction Of Soil Moisture Content

Soil moisture content directly affects the growth of crops. To accurately predict the amount of water in the soil and irrigating in proper times, we need to fully grasp the dynamic change of soil moisture content. But in practice, the change of soil water content is very complicated due to the influence of many factors, which has made the variation extremely complex, exhibiting nonlinear characteristics. This article based on the fractal theory, using the time series analysis method to study Xinjiang Fukang farmland soil moisture data, and forecasting the future trend of soil moisture content, therefore, some valuable conclusions have been obtained. Specific research steps are as follows:(1) This article uses R/S analysis to compare data time series of Hurst exponent with fractal dimension of comparative analysis of Fukang, Xinjiang farmland soil water content,the result turned to be that the soil water content has the variation characteristics of long term.The result of the research shows: the Hurst index of 10 cm, 80 cm, 150 cm are more than 0.5less than 1, we can get that the time series of soil water content in this area is a nonrandom time series and there is a long range correlation in the time series and the time series is persistent. The fractal dimension of time series in this region is greater than 1 and less than 2,showing that the variation of soil moisture content in the area gradually decreased with the increase of the depth.(2) This article used detrended fluctuation analysis method to study the long-range correlations time series of 14 different depth of soil moisture content in Fukang, Xinjiang, and using the fractal dimension to analysis the dynamic variation of soil moisture content. The result of the research shows: the Scaling exponent of time series of 14 different depth of soil moisture content increases with the increase of depth, which is showing that volatility changes of the soil moisture content increases with the increase of depth. The increase of fractal dimension has showed that the fluctuations of soil moisture content variation turned to be gradually stabilized with depth increased.(3) This article has cited three auto-regressive model to predict the oil moisture content in Fukang, Xinjiang, three kinds of the most ideal delay of time series were obtained. The results showed that: seasonal changes had a significant influence on the soil water content.This requires us that when we accurately predict soil moisture, we should pay more attention to the impact of seasonal variation in rainfall, so that we can obtain more accurate conclusions.