Application of Artificial Neural Networks, Support Vector Machine and Multiple Model-ANN to Sediment Yield Prediction
- PDF / 1,644,740 Bytes
- 15 Pages / 439.37 x 666.142 pts Page_size
- 73 Downloads / 250 Views
Application of Artificial Neural Networks, Support Vector Machine and Multiple Model-ANN to Sediment Yield Prediction Sarita Gajbhiye Meshram 1,2 Chandrashekhar Meshram 8
& Vijay P. Singh
3,4
& Ozgur Kisi
5,6
& Vahid Karimi
7
&
Received: 26 November 2019 / Accepted: 15 September 2020/ # Springer Nature B.V. 2020
Abstract
Sediment yield is important for maintaining soil health, reservoir sustainability, environmental pollution, and conservation of natural resources. The main aim of the present work is to develop four machine learning models, artificial neural networks (ANNs), radial basis function (RBF), support vector machine (SVM) and multiple model (MM)-ANNs for forecasting daily sediment yield. These models were applied to the Shakkar and Manot watersheds covering 25 years (1990–2015) and 10 years (2000–2010) of rainfall and discharge data, respectively. Results showed that the MM-ANNs model satisfactorily predicted sediment yield and outperformed the other models providing the highest correlation coefficient (0.921, 0.883) and Nash-Sutcliffe efficiency (0.744, 0.763) and the lowest relative absolute error (0.360, 0.344) and root mean square error (23,609.5, 269,671.5) for the Shakkar and Manot during the test period, respectively. Hence, the MM-ANNs model can be successfully used for sediment prediction. Keywords Machine learning models . Sediment yield . ANN . RBF . SVM . Multiple model
1 Introduction Watershed sediment load is an ecological hazard and its estimation is needed for developing measures for environmental protection, sustainability of reservoirs and hydropower generation, avoiding blockage of water supply systems, flood control, and maintaining soil fertility (Lin et al. 2006; Xu et al. 2012; Men et al. 2012). In many waterways, sediment is transported in suspension and estimation of suspended sediment (SS) is basic for designing channels, dams, and culverts (Targhi et al. 2017). Awareness of potential sediment loads is important for programmes for water resource
* Sarita Gajbhiye Meshram [email protected] Extended author information available on the last page of the article
Meshram S.G. et al.
management and environmental protection (Melesse et al. 2011). In fact, runoff and sediment yield models are regarded as the core components of the watershed planning and management tasks implemented through the concepts used for decision support by various resource managers. Soil erosion, which is directly related to issues with transporting sediments, continues to be a major ecological concern worldwide. Continuous monitoring of soil erosion and transport of sediments, however, can be a repetitive and highly demanding task; hence detailed models for forecasting these important decision-making parameters (Gajbhiye et al. 2014, 2015). In the last decades, artificial intelligence techniques have been applied in water resources management, especially for modelling processes with limited knowledge (Yoon et al. 2011; Chau and Wu 2010; Hsu et al. 1995). Such techniques include artificial
Data Loading...
