Suspended sediment load prediction using long short-term memory neural network

Abstract

Rivers carry suspended sediments along with their flow. These sediments deposit at different places depending on the discharge and course of the river. However, the deposition of these sediments impacts environmental health, agricultural activities, and portable water sources. Deposition of suspended sediments reduces the flow area, thus affecting the movement of aquatic lives and ultimately leading to the change of river course. Thus, the data of suspended sediments and their variation is crucial information for various authorities. Various authorities require the forecasted data of suspended sediments in the river to operate various hydraulic structures properly. Usually, the prediction of suspended sediment concentration (SSC) is challenging due to various factors, including site-related data, site-related modelling, lack of multiple observed factors used for prediction, and pattern complexity.Therefore, to address previous problems, this study proposes a Long Short Term Memory model to predict suspended sediments in Malaysia's Johor River utilizing only one observed factor, including discharge data. The data was collected for the period of 1988–1998. Four different models were tested, in this study, for the prediction of suspended sediments, which are: ElasticNet Linear Regression (L.R.), Multi-Layer Perceptron (MLP) neural network, Extreme Gradient Boosting, and Long Short-Term Memory. Predictions were analysed based on four different scenarios such as daily, weekly, 10-daily, and monthly. Performance evaluation stated that Long Short-Term Memory outperformed other models with the regression values of 92.01%, 96.56%, 96.71%, and 99.45% daily, weekly, 10-days, and monthly scenarios, respectively.

Artificial neural network model with different backpropagation algorithms and meteorological data for solar radiation prediction

“…The meteorological data collected are solely based on one meteorological station so this may pose a problem of the data being less diversified as the climatic condition is pretty constant. Malaysia is located near to the equatorial line, hence the tropical rainforest climate with high rate of rainfall 33 and overall high temperature throughout the entire year 34 is observed at the location of our studies. Besides, the meteorological data at night may also be captured by the meteorological station, hence resulting at the zero values of global radiation.…”

Predictive analysis of the value of information flow on the shop floor of developing countries using artificial neural network based deep learning

“…ANNs are adaptive to complex problems and systems even when the network typology changes; it can find complex relations among variables with good accuracy ( Ferrero Bermejo et al., 2019 ; Jalaee et al., 2019 ; Habtamu and Megersa, 2021 ). Following studies focusing on the comparative analysis ( Pandey et al., 2020 ; Aldahoul et al., 2021 ) of machine learning algorithms for regression such as linear regression (LR), DT, support vector machine (SVM), and ANN, we find that ANN is more efficient than the above algorithms. The complexity of the analysis of the information flow value will pass through an ANN model that will be built.…”

“… Pandey et al. (2020) and Aldahoul et al. (2021) discussed the performance of various machine learning algorithms for regression, such as decision trees (DT), multiple linear regression (MLR), and support vector machine (SVM), compared to artificial neural networks (ANNs).…”

Predicting Daily Suspended Sediment Load Using Machine Learning and NARX Hydro-Climatic Inputs in Semi-Arid Environment

“…Similarly, in hydrologic studies, ref. [21] applied a long short-term memory neural network (LSMNN) to predict the suspended sediment concentration (SSC) in the Johor River in Malaysia and found a high accuracy prediction in these models. In [22], three artificial intelligence models were used to estimate the sediment load in Ethiopia.…”