基于BMA多模型组合的疏勒河径流预测研究
周婷, 温小虎, 冯起, 尹振良, 杨林山

Study on runoff prediction of Shule River based on BMA multi-model combination
Ting ZHOU, Xiaohu WEN, Qi FENG, Zhenliang YIN, Linshan YANG
表3 SVM模型参数及预测结果评价
Table 3 Parameter and evaluation of prediction results of SVM model
模型参数训练集测试集
核函数参数(γ)惩罚因子(CRNSERMSE/(m3·s-1RNSERMSE/(m3·s-1
SVMt+1)0.25016.00.9710.91913.2230.9670.93514.435
SVMt+3)0.2500.50.8680.74624.1800.8860.75827.768
SVMt+5)0.004128.00.8110.64028.7920.8370.65533.157
SVMt+7)0.004256.00.7780.59530.5430.7830.58236.494