Rainfall prediction in coastal hilly areas based on VMD–RSA–DNC DOI Creative Commons
Xianqi Zhang, Qiuwen Yin, Fang Liu

et al.

Water Science & Technology Water Supply, Journal Year: 2023, Volume and Issue: 23(8), P. 3359 - 3376

Published: July 31, 2023

Abstract Highly accurate rainfall prediction can provide a reliable scientific basis for human production and life. For the characteristics of occasional sudden changes in coastal hilly areas, this article chooses four cities eastern Zhejiang province as object study establishes model based on variational mode decomposition (VMD), reptile search algorithm (RSA), differentiable neural computer (DNC). The VMD reduces complexity sequence data; RSA is used to find best-fit function; DNC combines advantages recurrent network computational processing improve problem memory forgetting long short-term memory. To verify accuracy model, results are compared with other three models, show that VMD–RSA–DNC has best maximum minimum relative errors 9.62 0.17%, respectively, average root-mean-square error 5.43, mean absolute percentage 3.59%, Nash–Sutcliffe efficiency 0.95 predicting area. This provides new reference method construction models.

Language: Английский

Predicting the flood peak arrival time via a comprehensive machine learning framework: case studies in Changhua and Tunxi basins, China DOI Creative Commons
Shi Zhou, Xiaona Liu

Journal of Water and Climate Change, Journal Year: 2024, Volume and Issue: 16(1), P. 142 - 159

Published: Dec. 10, 2024

ABSTRACT Floods are becoming increasingly frequent and severe due to climate change urbanization, thereby increasing risks lives, property, the environment. This necessitates development of precise flood forecasting systems. study addresses critical task predicting peak arrival times, which is essential for timely warnings preparations, by introducing a comprehensive machine-learning framework. Our approach integrates interpretable feature engineering, individual model design, novel ensembles enhance prediction accuracy. We extract informative features from historical flow rainfall data, design suite models, develop ensemble technique combine predictions. conducted case studies on Tunxi Changhua basins in China. Numerical experiments reveal that our method significantly benefits engineering ensembles, achieving mean absolute error (MAE) errors 1.524 h 2.192 Changhua. These results notably outperform best baseline method, achieves MAE 1.727 2.737

Language: Английский

Citations

0
Four-channel generative adversarial networks can predict the distribution of reef-associated fish in the South and East China Seas DOI
Jia Wang, Shigeru Tabeta

Ecological Informatics, Journal Year: 2023, Volume and Issue: 78, P. 102321 - 102321

Published: Sept. 30, 2023

Language: Английский

Citations

1
Hybrid model-based prediction of biomass density in case studies in Turkiye DOI Creative Commons

B. İşler,

Zafer Aslan, F. Sunar

et al.

Ecological Informatics, Journal Year: 2023, Volume and Issue: 79, P. 102439 - 102439

Published: Dec. 16, 2023

Growing global concern over natural resource degradation due to urbanisation and population growth emphasizes the critical need for innovative solutions. Addressing this imperative, our study pioneers integration of cutting-edge artificial intelligence (AI) methods investigate crucial changes in vegetation density. In context, a hybrid model, which harmoniously integrates conventional neural network (ANN) models with Wavelet-ANN (W-ANN) approach, was employed two case pilot areas, namely on Alanya Antalya Iznik Bursa, Turkiye, renowned their distinct ecosystems land cover patterns. By employing diverse data sources, encompassing satellite-derived metrics such as Enhanced Vegetation Index (EVI) Land Surface Temperature (LST) from MODIS/Terra satellite, alongside atmospheric data, investigation intricately temporal dynamics extending year 2030. Remarkably, W-ANN model demonstrates better predictive performance compared methodologies. It anticipates substantial 21.4% reduction biomass density Iznik, achieving minimal 5.4% error probability. Similarly, Alanya, forecasts notable 6.6% decrease remarkably low 2% probability, both projections Our reveals significant by comparing projected values 2030 observed 2018. These findings gain further support an analysis Normalised Difference Built-up (NDBI) derived Landsat satellites, affirming exceptional efficacy AI-driven approach advancing understanding urbanisation's impact ecosystems.

Language: Английский

Citations

1
Rainfall prediction in coastal hilly areas based on VMD–RSA–DNC DOI Creative Commons
Xianqi Zhang, Qiuwen Yin, Fang Liu

et al.

Water Science & Technology Water Supply, Journal Year: 2023, Volume and Issue: 23(8), P. 3359 - 3376

Published: July 31, 2023

Abstract Highly accurate rainfall prediction can provide a reliable scientific basis for human production and life. For the characteristics of occasional sudden changes in coastal hilly areas, this article chooses four cities eastern Zhejiang province as object study establishes model based on variational mode decomposition (VMD), reptile search algorithm (RSA), differentiable neural computer (DNC). The VMD reduces complexity sequence data; RSA is used to find best-fit function; DNC combines advantages recurrent network computational processing improve problem memory forgetting long short-term memory. To verify accuracy model, results are compared with other three models, show that VMD–RSA–DNC has best maximum minimum relative errors 9.62 0.17%, respectively, average root-mean-square error 5.43, mean absolute percentage 3.59%, Nash–Sutcliffe efficiency 0.95 predicting area. This provides new reference method construction models.

Language: Английский

Citations

0