A kind of soft sensing method for biomass concentration of phytoplankton in seawater

Effective monitoring the growth state of seawater phytoplankton plays an important role for the early warning of marine disasters, such as coastal red tides. Grey correlation analysis method was used to select the secondary variables of the soft sensing model. It can effectively reduce the dimension of the system. Extreme learning machine regression (ELMR) method was used to build the soft sensing model of biomass concentration of phytoplankton. Comparing with the generalized regression neural network, the testing result indicates that extreme learning machine regression has better accuracy, efficiency and generalization ability of measurement than the other methods. It adapts to be used for real time monitoring of biomass concentration of phytoplankton in seawater.