Author(s): Ah Dogan; Shivam Tripathi; D. A. Lyn; Rao S. Govindaraju
Linked Author(s):
Keywords: Relevance vector machine; Sediment concentration; Total sediment load; Sediment transport equations
Abstract: The transport of sediment in rivers affects channel navigability, reservoir filling, hydroelectric-equipment longevity, fish habitats, river pollution, and river aesthetics. Sediment concentrations may be determined from direct measurements, or estimated from sediment transport equations that require detailed information about the flow and sediment characteristics. However, there is often a large discrepancy between model predictions and observations because complexities of sediment transport phenomena are not well understood, thereby motivating the search for alternate methods. As a fairly recent computing tool, relevance vector machines (RVMs) are gaining popularity in the fields of machine learning and pattern recognition. The main purpose of this study is to examine an RVM approach for estimation of total sediment load. The effectiveness of the proposed approach is illustrated through its application to more than 1000 laboratory data sets compiled by Brownlie (1981b). Dimensional analysis is used for finding input parameters for RVM. The proposed RVM approach is shown to be statistically superior to other well known sediment transport equations.
Year: 2007