基于支持向量机的水资源预测模型

Prediction Model of Water Resources based on Support Vector Machine

水资源信息的预测可以为水资源的合理调配、管理和规划提供依据,为提高拟合和预测精度,建立基于支持向量机的水资源信息预测模型,采用捕鱼算法对其参数优化,并应用于地下水水位预测的实例分析,且拟合及预测精度均较高,与神经网络模型方法所取得的结果进行比较,能取得更好效果,表明将回归支持向量机用于水资源信息的拟合、预测是可行的。

Prediction of water resources information provides the basis for reasonable allocation, management and planning of water resources. In order to improve the fitting and prediction accuracy, this paper establishes a predic- tion model of water resources information based on support vector machine, optimizes its parameter by Optimization algorithm on using fishing strategy （FSOA）, applies it to the example analysis of groundwater level prediction, and finds that both fitting and prediction accuracy are higher, which can get better result, compared to the result obtained from Neural network model method. It shows that it is feasible for regression support vector machine to be used for fitting and prediction of water resources information.