Search

Ying, J.Y.; Liang, K.W.; Wu, Q.S.; Xie, M.; Jin, X.C.; Ye, Q., and Yang, Z.L., 2020. Calculation of suspended sediment concentration based on deep learning and OBS turbidity. In: Bai, X. and Zhou, H. (eds.), Advances in Water Resources, Environmental Protection, and Sustainable Development. Journal of Coastal Research, Special Issue No. 115, pp. 627-630. Coconut Creek (Florida), ISSN 0749-0208. Based on BP and Elman deep learning models with water depth, velocity, flow direction and salinity as input term and suspended sediment concentration as output term were constructed, and the calculated results were compared with the measured suspended sediment concentration and the suspended sediment concentration calculated by OBS turbidity. The results show that the suspended sediment concentration calculated by deep learning model can meet the needs of sediment dynamics research, but the calculation effect is not so good in high water stand, low water stand, fastest flood and fastest ebb periods, and the accuracy is far less than that of OSB calculation results. In the future, deep learning models can be improved in computational accuracy by adding input terms, and experimentally adjusting thresholds and connection weights. [ABSTRACT FROM AUTHOR]

Ying, J.Y.; Liang, K.W.; Wu, Q.S.; Xie, M.; Jin, X.C.; Ye, Q., and Yang, Z.L., 2020. Calculation of suspended sediment concentration based on deep learning and OBS turbidity. In: Bai, X. and Zhou, H. (eds.), Advances in Water Resources, Environmental Protection, and Sustainable Development. Journal of Coastal Research, Special Issue No. 115, pp. 627-630. Coconut Creek (Florida), ISSN 0749-0208. Based on BP and Elman deep learning models with water depth, velocity, flow direction and salinity as input term and suspended sediment concentration as output term were constructed, and the calculated results were compared with the measured suspended sediment concentration and the suspended sediment concentration calculated by OBS turbidity. The results show that the suspended sediment concentration calculated by deep learning model can meet the needs of sediment dynamics research, but the calculation effect is not so good in high water stand, low water stand, fastest flood and fastest ebb periods, and the accuracy is far less than that of OSB calculation results. In the future, deep learning models can be improved in computational accuracy by adding input terms, and experimentally adjusting thresholds and connection weights. [ABSTRACT FROM AUTHOR]