Water-Bloom Medium-Term Prediction Based on Gray-BP Neural Network Method

On the basis of studying the mechanism of water bloom, one kind of gray-BP artificial neural network forecasting method is proposed in the paper. The gray theory was used to obtain preliminary forecast of the occurrence trend of water bloom, combined with neural network to implement error compensation for the forecast result. Compared with BP, this method can predict chlorophyll change trend more accurately, and significantly improve the prediction accuracy with the prolongation of prediction period. It provides an effective new method for water bloom medium-term prediction.