BP神经网络在2005年太湖蓝藻水华预警中的应用

蓝藻水华是目前中国乃至世界面临的重大环境问题之一。为了有效地减少及预防蓝藻水华带来的影响，收集苏州市吴县1986—2007年的气象资料，运用主成分分析法，分析了太湖蓝藻暴发前一个月的主要限制因子及其相互关系。气温、气压、相对湿度、降水是影响叶绿素a浓度的主要限制因子。结合2005年1—10月太湖各区域蓝藻叶绿素a浓度的含量，利用Matlab R2010a软件，建立了基于BP神经网络的蓝藻水华预警模型，可为采取相应措施和控制蓝藻水华提供科学依据。

The Application of BP Neural Network in Early Warning of Algal Blooms in Year 2005 in Lake Taihu

Algal blooms is one of the serious environmental problems in our country and even the world at present. In order to reduce and prevent the effects of algal blooms effectively, it collects the meteorological data in meteorological stations of Wuxian, Suzhou from 1986 to 2007, and uses principal component analysis to analyze the main limiting factors and their interrelation one month before the outbreak of algae in Lake Taihu. The average temperature, average pressure, average relative humidity and cumulative precipitation are the main limiting factors in influencing chlorophylla. Combined with the concentration of chlorophyll a from January to October in 2005 in various regions in Lake Taihu, Used the software of Matlab R2010a, the early warning model of algal blooms was established based on BP neural network, this provides scientific basis in order to take appropriate measures to control algal blooms.