Artificial neural network application in sediment prediction: a review
The development of an accurate sediment prediction model is a priority for all hydrological researchers. Several conventional approaches showed an inability to predict suspended sediment correctly. These approaches fall short in understanding the transport of sediments behaviour in rivers because of...
| Main Authors: | , , , , , |
|---|---|
| Other Authors: | |
| Format: | Book Section |
| Language: | English |
| Published: |
Penerbit UTHM
2020
|
| Subjects: | |
| Online Access: | http://eprints.uthm.edu.my/2711/ http://eprints.uthm.edu.my/2711/1/Ch03.pdf |
| Summary: | The development of an accurate sediment prediction model is a priority for all hydrological researchers. Several conventional approaches showed an inability to predict suspended sediment correctly. These approaches fall short in understanding the transport of sediments behaviour in rivers because of its complex, non-stationary, and non-linear nature. There have been important advances in the theoretical understanding of Artificial Neural Network (ANN) over the last few decades, as well as algorithmic techniques for their implementation and applications of the approach to hydrological and practical problems. ANN and other machine learning models have been used in predicting complex non-linear relationships and patterns of large input parameters to achieve the desired output. This chapter reviews several relevant works of literature on sediment transport prediction while focusing on a wide range of ANN applications. ANN sediment transport models have increasingly attracted several researchers over the past years. Therefore, the need to acquire in-depth knowledge about their theory and modelling approaches. Besides, this chapter provides an overview of the ANN approach and other emerging machine learning hybrid models, which have yielded satisfactory results. Also, provided in this review are several examples of successful applicability of ANN in sediment prediction. |
|---|