YUAN Hongchun, MAO Rui, YANG Mengzhao, et al. Research on dissolved oxygen prediction model based on XGBoost[J]. 2020, (10): 51-53+57. DOI: 10.13873/J.1000-9787(2020)10-0051-03.
In order to improve precision of dissolved oxygen prediction
XGBoost model is applied to predict the water quality parameters
on the basis of traceability of aquatic products and monitoring data safety warning platform. And compared with the LSTM
PCA-NARX network prediction model for 48 hours of dissolved oxygen prediction. Simulation results show that the root mean square error( RMSE) of the results predicted by the XGBoost model is the smallest
and the generalization ability in 24 and 48 hours is higher than that of the LSTM and PCANARX models.