论文部分内容阅读
准确预测干旱区地下水埋深,对区域地下水资源的合理开发利用与生态环境保护具有十分重要的意义。以额济纳盆地3个地下水埋深观测井为对象,运用小波变换与支持向量机耦合模型(WA-SVM)对观测井未来1个月的地下水埋深进行了短期预测。为检验WA-SVM的有效性,将模拟结果与未经小波变换的SVM模型进行了对比。结果表明:在对干旱区地下水埋深进行短期预测时,相较于SVM模型,WA-SVM模型的预测精度显著提高。WA-SVM模型在干旱区地下水埋深预测中有更好的适用性,可以为干旱地区地下水埋深动态预测提供新的方法和思路,是资料有限的条件下地下水埋深预测的有效方法。
Accurately predicting the groundwater depth in arid areas is of great significance to the rational exploitation and utilization of regional groundwater resources and ecological environment protection. Taking three groundwater depth observation wells in the Ejin basin as an example, short-term prediction of the groundwater depth in the next one month of the observation well is carried out by using the wavelet transform and support vector machine coupled model (WA-SVM). To test the effectiveness of WA-SVM, the simulation results are compared with the non-wavelet-based SVM model. The results show that the prediction accuracy of WA-SVM model is significantly higher than that of SVM model when short-term prediction of groundwater depth in arid area. WA-SVM model has better applicability in predicting groundwater depth in arid area, which can provide a new method and thought for dynamic prediction of groundwater depth in arid area. It is an effective method to predict groundwater depth under limited data.