论文部分内容阅读
三江平原是我国粮食主产区之一,近年来农业水资源出现危机.预测该地区地下水动态变化趋势,对于指导该地区合理开发利用地下水有着重大的理论和现实意义.建立了基于RAGA的灰色BP神经网络预测模型.该模型克服了传统GM(1,1)模型存在明显系统误差的缺点,既具有GM(1,1)模型对数据确定性方面把握的长处,也融合了人工神经网络在不确定因素预测领域的优势.通过两种途径进行检验,结果表明该模型具有相对较高的预测精度.运用该模型对三江平原地下水埋深进行动态预测,未来五年内,如果仍按目前的发展模式,该地区地下水埋深仍将持续下降,从2007年到2012年,该地区地下水平均每年下降0.3m.预测结果对有关部门的政策决策具有一定的指导意义.
Sanjiang Plain is one of the main grain-producing areas in our country and the crisis of agricultural water resources has occurred in recent years.It is of great theoretical and practical significance to predict the dynamic change of groundwater in this area to guide the rational exploitation and utilization of groundwater in this area.Graph BP This model overcomes the shortcoming that the traditional GM (1,1) model has obvious systematic errors. It not only has the advantages of the GM (1,1) model in determining certainty of the data, but also combines the advantages of the artificial neural network Determine the predominance of factors in the field of forecasting.Through two ways to test, the results show that the model has a relatively high prediction accuracy.Using the model to predict the groundwater depth in the Sanjiang Plain dynamic prediction, the next five years, if still according to the current development model , The groundwater depth in this area will continue to decline, and the average groundwater level in this area will drop 0.3m annually from 2007 to 2012. The forecast results will be of guiding significance to the policy-making of relevant departments.