Study On Impact Factors And Methods Of Return Water Volume Prediction In Yellow River Irrigation Area Of Ning Xia

Yellow River irrigated area of Ningxia is one of China's large ancient irrigation, The return water volume typically accounts for 40%-60% of of annual diversion. Its re-use not only for irrigation area, but also for the entire Yellow River water regulation has great significance.Through reviewing domestic and foreign relevant literature, utilizing the approach combining the theoretical methods and application, the paper analyses the variation regularity of return water of Yellow River irrigated area of Ningxia, ascertains the composition and main influencing factors of return water, classifies the return water into three different types:annual return water, monthly return water and daily return water, and researches the theory and method of predicting return water of irrigation area. Among the key findings:1.The paper reveals the variation regularity of return water in multi-years and single-year, and ascertains the main influencing factors of return water using methods of the grey relational degree analysis and correlation analysis,they are diversion, precipitation and groundwater table.2. Studing on the year return water volume use the Research RBF network model, with an annual return water prediction models. Research shows that the Internet can be a very good approximation of training samples, test sample predictive accuracy.3. Changes in irrigation water was back on a certain degree of periodicity, but the random fluctuations, it is difficult with conventional numerical analysis method to simulate, this paper, "Decomposition-reconstruction-prediction" of the wavelet network model, wavelet decomposition of the irrigation district On back water sequences were smooth and steady decomposition of the sequence after the completion of the original signal is good, and implements the right characteristics of the wavelet coefficients of different sequences select a different neural network model to predict and improve the prediction accuracy.4. The sequence of irrigation water on back of non-stationary time series, this paper, wavelet time series forecasting methods, namely, the use of wavelet analysis to time series can be decomposed by wavelet decomposition level by level, to a different frequency channel, due to decomposition of the sequence of in the frequency components than the original signal a single, and made a smooth wavelet decomposition of signals, therefore, decomposed the signal to the smooth nature of much better than the original signal, the wavelet decomposition in order to ensure an adequate sample sequence number on back of water for irrigation Time series decomposition and then reconstructed to the original scale, right after the wavelet decomposition and reconstruction sequences using the corresponding time series model for prediction.5. By carrying chaotic identification and phase space reconstruction on daily return water of irrigation area. The papers finds the chaotic character of daily return water of irrigation area, ascertains the time delay, embedment dimension and correlation dimension of phase space reconstruction, on this basis, the phase space reconstruction theory and wavelet network method for combining the establishment of irrigation water withdrawal forecasts on wavelet network model for chaotic, overcoming the traditional BP network vulnerable to local minima, slow convergence and other shortcomings.