On state prediction for algae growth in seawater based on fuzzy back-propagation network

The state of algae reproduction is a key index for the status of seawater quality and pollutants emission for rivers. Algae growth is affected by many physical-chemical factors, this kind of complex relationship is difficult to be described by ordinary mechanism expression. Fuzzy back-propagation network can describe the complex nonlinear system, and it has a fine performance of generalization, it can give a dynamic estimate to the output variables of the system. We use PCA(Principal Component Analysis) method to reduce the dimension of the sample data, simplify the complexity of the model system, it can make the model has a fine convergence rate. The practical testing illustrates that fuzzy back-propagation network model based on PCA can be applied in state prediction for algae growth to good purpose.