Wan Hanna Melini Wan Mohtar

ANN Based Sediment Prediction Model Utilizing Different Input Scenarios

Modeling sediment load is a significant factor in water resources engineering as it affects directly the design and management of water resources. In this study, artificial neural networks (ANNs) are employed to estimate the daily sediment load. Two different ANN algorithms, the feed forward neural network (FFNN) and radial basis function (RBF) have been used for this purpose. The neural networks are trained and tested using daily sediment and flow data from Rantau Panjang station on Johor River. The results show that combining flow data with sediment load data gives an accurate model to predict sediment load. The comparison of the results indicate that the FFNN model has superior performance than the RB model in estimating daily sediment load.

RBFNN versus FFNN for daily river flow forecasting at Johor River, Malaysia

Abstract Streamflow forecasting can have a significant economic impact, as this can help in water resources management and in providing protection from water scarcities and possible flood damage. Artificial neural network (ANN) had been successfully used as a tool to model various nonlinear relations, and the method is appropriate for modeling the complex nature of hydrological systems. They are relatively fast and flexible and are able to extract the relation between the inputs and outputs of a process without knowledge of the underlying physics. In this study, two types of ANN, namely feed-forward back-propagation neural network (FFNN) and radial basis function neural network (RBFNN), have been examined. Those models were developed for daily streamflow forecasting at Johor River, Malaysia, for the period (1999–2008). Comprehensive comparison analyses were carried out to evaluate the performance of the proposed static neural networks. The results demonstrate that RBFNN model is superior to the FFNN forecasting model, and RBFNN can be successfully applied and provides high accuracy and reliability for daily streamflow forecasting.

Application of the Hybrid Artificial Neural Network Coupled with Rolling Mechanism and Grey Model Algorithms for Streamflow Forecasting Over Multiple Time Horizons

Streamflow forecasting is paramount process in water and flood management, determination of river water flow potentials, environmental flow analysis, agricultural practices and hydro-power generation. However, the dynamicity, stochasticity and inherent complexities present in the temporal evolution of streamflow could hinder the accurate and reliable forecasting of this important hydrological parameter. In this study, the uncertainty and nonstationary characteristics of streamflow data has been treated using a set of coupled data pre-processing methods before being considered as input for an artificial neural network algorithm namely; rolling mechanism (RM) and grey models (GM). The rolling mechanism method is applied to smooth out the dataset based on the antecedent values of the model inputs before being applied to the GM algorithm. The optimization of the input datasets selection was performed using auto-correlation (ACF) and partial auto-correlation (PACF) functions. The pre-processed data was then integrated with two artificial neural network models, the back propagation (RMGM-BP) and Elman Recurrent Neural Network (RMGM-ERNN). The development, training, testing and evaluation of the proposed hybrid models were undertaken using streamflow data for two tropical hydrological basins (Johor and Kelantan Rivers). The hybrid RMGM-ERNN was found to provide better results than the hybrid RMGM-BP model. Relatively good performance of the proposed hybrid models with a data pre-processing approach provides a successful alternative to achieve better accuracy in streamflow forecasting compared to the traditional artificial neural network approach without a data pre-processing scheme.

Optimisation of Multiple Hydropower Reservoir Operation Using Artificial Bee Colony Algorithm

In this study, the Artificial Bee Colony (ABC) algorithm was developed to solve the Chenderoh Reservoir operation optimisation problem which located in the state of Perak, Malaysia. The proposed algorithm aimed to minimise the water deficit in the operating system and examine its performance impact based on monthly and weekly data input. Due to its capability to identify different possible events occurring in the reservoir, the ABC algorithm provides promising and comparable solutions for optimum release curves. The optimal release curves were then used to stimulate the reservoir release under different operating times under different inflow scenarios. To investigate the performance of both the monthly and weekly ABC optimisation employed in the reservoir, the well-known reliability, resilience and vulnerability indices were used for performance assessment. The indices tests revealed that weekly ABC optimisation outperformed in terms of reliability and vulnerability leading to the development of a better release policy for optimal operation.

Oscillating-grid turbulence at large strokes: Revisiting the equation of Hopfinger and Toly

A quasi-isotropic, quasi-homogeneous turbulence generated by an oscillating-grid, spatially decays according to power law of u∝ Z−−nu, where u is the root mean square (rms) horizontal velocity, Z is the vertical distance from the grid and nu =1. However, the findings of Nokes and Yi indicate that as the stroke of oscillation increases, the power law nu ≠ 1 and does not follow the established decay law equation of Hopfinger. This paper investigates the characteristics of the turbulence that are generated using larger strokes S/M = 1.6 and 2 and compares with that obtained using a S/M = 0.8, which is the stroke used when the equation was developed. Measurements of the grid-generated turbulence in a water tank were taken using particle image velocimetry (PIV). The results showed that the homogeneity occurred at distance beyond 2.5 mesh spacings away from the grid midplane, independent of the stroke and the frequency of oscillation. Within this region, the turbulent kinetic energy distribution was quasi-homogeneous, and the secondary mean flow is negligible. The statistical characteristics of the measured turbulence confirmed that although nu decreases as stroke increases, the grid-turbulence generated at S/M = 1.6 and 2 obeys the universal decay law.

Textural characteristics and sedimentary environment of sediment at eroded and deposited regions in the severely eroded coastline of Batu Pahat, Malaysia

This study investigates the textural characteristics of sediments collected at eroded and deposited areas of highly severed eroded coastline of Batu Pahat, Malaysia. Samples were taken from systematically selected 23 locations along the 67km stretch of coastline and are extended to the fluvial sediments of the main river of Batu Pahat. Grain size distribution analysis was conducted to identify its textural characteristics and associated sedimentary transport behaviours. Sediments obtained along the coastline were fine-grained material with averaged mean size of 7.25 ϕ, poorly sorted, positively skewed and has wide distributions. Samples from eroded and deposition regions displayed no distinctive characteristics and exhibited similar profiles. The high energy condition transported the sediments as suspension, mostly as pelagic and the sediments were deposited as shallow marine and agitated deposits. The fluvial sediments of up to 3km into the river have particularly similar profile of textural characteristics with the neighbouring marine sediments from the river mouth. Profiles were similar with marine sediments about 3km opposite the main current and can go up to 10km along the current of Malacca Straits.

Influence of bed deposit in the prediction of incipient sediment motion in sewers using artificial neural networks

ABSTRACT This study investigates the performance of artificial neural networks in predicting the incipient sediment motion in sewers. Two neural network algorithms, i.e. feed forward neural network (FFNN) and radial basis function (RBF), were employed to estimate the critical velocity over varying sediment thickness, median grain size and water depth. Empirical data from five studies were fed into the models and the performance of each model was scrutinized based on three performance criteria. Prediction from FFNN was found to give higher accuracy than values obtained from RBF. Analysis was also extended to observe the correlation between the predicted critical velocity with calculated critical velocity using five empirical equations developed using non-linear regression analysis. Prediction by FFNN proved to have the highest accuracy compared to the RBF and the values obtained through empirical equations described in this study.

The development of automated spatial and temporal measurement system for lab-scale local scour

The spatial and temporal measurements of local scour around bridge piers provide the quantification of local scouring process. Most studies on a laboratory scale faced difficulties in obtaining a holistic local spatial variability around bridge pier, especially for continuous small time interval. Long experimental period, which could take up to few days, does not permit a consistent time interval spatial scour measurement due in particular to the physical constraints that exist under laboratory conditions. This study proposed an automated, cost-effective system which is capable of detecting changes in both spatial and temporal local scour. The system allows measurement to be made by using data recorder at an adjustable distance (± 0.1 mm) and angle (± 0.1˚) from the original position, which is programmed and controlled with an Arduino, which is an open source microcontroller with multiple capability of controlling electrical components such as motors and sensors. When the data recorder is in position, data is automatically captured and sequentially saved at a particular spatial interval. In this study, a web camera was used as a data recorder to capture images in the azimuthal plane for a one-hour interval. Images were captured for 30 seconds per measurement per position. The system was set up to monitor and measure the temporal and spatial local scour continuously in an 80-hour experiment. Results show that the location of maximum scour depth varied for different time intervals, and migrated from downstream to upstream of the pier. The rate of scour decreased as duration of experiment was increased. The system was able to provide a holistic view of both spatial and temporal variability in the development of local scour on a laboratory scale.

Characteristics of Low Reynolds Number Shear-Free Turbulence at an Impermeable Base

Shear-free turbulence generated from an oscillating grid in a water tank impinging on an impermeable surface at varying Reynolds number 74 ≤ Re l ≤ 570 was studied experimentally, where the Reynolds number is defined based on the root-mean-square (r.m.s) horizontal velocity and the integral length scale. A particular focus was paid to the turbulence characteristics for low Re l < 150 to investigate the minimum limit of Re l obeying the profiles of rapid distortion theory. The measurements taken at near base included the r.m.s turbulent velocities, evolution of isotropy, integral length scales, and energy spectra. Statistical analysis of the velocity data showed that the anisotropic turbulence structure follows the theory for flows with Re l ≥ 117. At low Re l < 117, however, the turbulence profile deviated from the prediction where no amplification of horizontal velocity components was observed and the vertical velocity components were seen to be constant towards the tank base. Both velocity components sharply decreased towards zero at a distance of ≈1/3 of the integral length scale above the base due to viscous damping. The lower limit where Re l obeys the standard profile was found to be within the range 114 ≤ Re l ≤ 116.