基于信息扩散近似推理的年径流量预测模型

Prediction Model of Annual Runoff Based on Information Diffusion Approximate Reasoning

通过分析年径流时间序列的特性,利用信息扩散近似推理描述年径流量间的复杂非线性关系,建立起基于信息扩散近似推理的年径流预测模型。信息扩散近似推理将样本点转换成模糊集,部分弥补了由于数据的不完备性所造成的信息空白,并可以将矛盾模式转换成兼容模式。通过与传统预测方法相比较,发现该模型能够很好地光滑样本数据以及能够较好地发掘知识,有较高的预测精度和推广应用价值。

According to the features of annual runoff time series,the prediction model of annual runoff based on information diffusion approximate reasoning is established.Runoff regulations are described by information diffusion approximate reasoning that can take advantage of more information.Results indicate that annual runoff prediction with information diffusion approximate reasoning model is good at mining uncertain knowledge and can find out more information than traditional methods.