Pelajar UiTM menang tesis terbaik 2018 di Majlis Perasmian Sambutan Air Sedunia Peringkat Kebangsaan

Tarikh: 24 Mac 2018

Masa: 8.00 pagi - 12.00 tengahari

Tempat: Laman Budaya Kuala Kangsar, Perak

Tajuk Tesis: Streambank Erosion Prediction Using Empirical Model for Natural River Channels

Supervisor(s): Prof. Dr. Ir. Hajah Junaidah Ariffin & Dr Jazuri Abdullah.

Streambank Erosion Prediction Using Empirical Model for Natural River Channels

Azlinda Saadon, PhD. in Civil Engineering, UiTM.

This thesis has been conducted by Azlinda Saadon, a student from Universiti Teknologi Mara (UiTM) entitled Streambank Erosion Prediction using Empirical Model for Natural River Channels. This thesis was supervised by Prof. Dr. Ir. Hajah Junaidah Ariffin and Dr Jazuri Abdullah whose field of interest are in hydraulics and river engineering. Azlinda is currently attached at The Infrastructure University Kuala Lumpur (IUKL) as a fulltime lecturer and she is also holding a post as Coordinator (Training and Publication Unit) for I-GEO Disaster Research Centre, IUKL. Her passion and interest towards river and water influenced her to conduct research on the riverbank erosion.

THESIS SUMMARY

Streambank erosion is one of the complex problems in river engineering studies as it requires integration from various fields of engineering. It is commonly associated with river meandering initiation and development, river width adjustment and river plan-form evolution. It requires integration between soil and water to properly understand the factors that constitutes to streambank erosion and its impact to major scouring. This study was undertaken to investigate the factors of streambank erosion and to quantify the rates of streambank erosion at the areas susceptible to erosion. Fieldwork investigation technique was conducted in the quantification of streambank erosion rates. Erosion pin arrays and streambank vertical profiling techniques were used in the quantification of streambank erosion rates. Dimensional analysis was performed to establish the factors governing the rates of streambank erosion. The functional relationship derived from dimensional analysis emphasized that the governing factors to streambank erosion includes flow-induced factors, resistance to the soil, streambank geometry, and sediment grain resistance. Selection of the most significant parameters constitutes to streambank erosion rates is obtained from the analysis. The study includes the development of the newly streambank erosion expression using three techniques, namely, Statistical Approach, Nonlinear Autoregressive Exogenous (NARX) model, and Artificial Neural Network (ANN). Result analyses and model validation show that Artificial Neural Network (ANN) performed in good agreement with the measured data. The model produced very good prediction up to 90% accuracy. Statistical model and NARX model predicted equally good performance ranging from 70% - 90% accuracy. The empirical models in predicting streambank erosion rates developed in this study will contribute as a valuable tool and good guidance for supporting streambank monitoring at the areas susceptible to erosion. Establishment of reliable predictive models with regards to streambank erosion can minimize the impact due to scouring at existing infrastructures such as bridge piers or abutments, which further lead to landslides and river bank collapse.