变权组合模型在景观水体水质模拟中的应用

Application of combination model of variable weight in landscape water quality simulation

针对景观水体的水质模拟与预测问题,在BP神经网络和支持向量机模型的基础上,建立了权重随输入量变化的变权组合模型。该模型既能充分利用各个单一模型的优点,又能避免固定权重分配的弊端。经实例验证,与单一的BP神经网络和支持向量机模型相比,变权组合模型拟合精度更高,预测结果更为准确。

In order to simulate and predict landscape water quality,we proposed a combination model of variable weight,which based on the back-propagation（ BP） neural network model and support vector machine model. The combination model of variable weight could make full use of the single model,and it would avoid the drawbacks of the fixed weight distribution. With the case study and comparison of the BP neural network model and support vector machine mode,we could find that the combination model of variable weight is much better in fitting accuracy and prediction results.