摘要
利用误差反向传播(BP)人工神经网络模型建立了天津赤潮驴驹河赤潮监控区各种理化因子与叶绿素-a之间的映射关系。模拟结果表明,该模型能够较好的反映各种理化因子与叶绿素-a浓度之间的非线性对应关系,可间接地用于该地区的赤潮预测。
By using BP Neural Network method, a nonlinear relationship between various physical-chemical factors and the chlorophyll-a concentration in the Bohai Gulf is established. The simulated results show that the predicted chlorophyll-a concentration fits the observed data well, indicating that this model can be indirectly used for local red tide prediction.
出处
《海洋预报》
2008年第2期1-10,共10页
Marine Forecasts
基金
国家"十五"科技支撑项目(2006BAV03B01)
