A combined optimization prediction model for earth-rock dam seepage pressure using multi-machine learning fusion with decomposition data-driven

Liu, Bo; Cen, Weijun; Zheng, Changhai; Li, Dengjun; Wang, Libo

doi:10.1016/j.eswa.2023.122798

Cited by 7 publications

(1 citation statement)

References 75 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Dams play a crucial role in water resource management, irrigation, power generation, and flood control [1]. The highquality, efficient, and safe construction and long-term efficient and safe operation of dams are crucial to national welfare and people's livelihoods [2].…”

Section: Introductionmentioning

confidence: 99%

Prediction of dike seepage pressure based on ISSA-BiLSTM

Chen,

Liu,

Xing

et al. 2024

Preprint

View full text Add to dashboard Cite

Seepage behavior is one of the critical factor in the operational safety of dams, and predicting dam seepage is the key content in dam monitoring and safety assessment research. The existing traditional dam seepage pressure prediction models have problems such as easy to fall into local optimum and limited predictive efficiency. The sparrow search algorithm(SSA) was improved as ISSA using both methods nonlinear Sine Cosine optimization algorithm and adaptive producer and scrounger ratio. We combined the Bidirectional Long Short-Term Memory (BiLSTM) neural network model with ISSA to develop the ISSA-BiLSTM seepage pressure prediction model. And the critical feature factors were extracted based on LightGBM to construct the input layer for seepage pressure prediction. The results show that the ISSA-BiLSTM model's fitting outcomes are generally consistent with the observed changes in seepage pressure observations, achieving an R2 of 0.987. In comparison to SSA-BiLSTM and BiLSTM, the model exhibits a substantial reduction in errors, decreasing by approximately 20% and 30%, respectively. This model can provide technical support and insights for accurately predicting dam seepage, contributing to the advancement of this field.

show abstract

Section: Introductionmentioning

confidence: 99%