Development of riverbank erosion rate predictor for natural channels using NARX-QR Factorization model: a case study of Sg. Bernam, Selangor, Malaysia

Saadon, Azlinda; Abdullah, Jazuri; Muhammad, Nur Shazwani; Ariffin, Junaidah

doi:10.1007/s00521-020-04835-5

Cited by 12 publications

(3 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…This study emphasizes the significance of the data population in developing prediction models employing mathematical, statistical, and artificial intelligence systems (Saadon et al, 2021;Saadon et al, 2020;Dobbin and Simon, 2011). The data splitting approach used in this research involved optimal split proportions, where 60% of the data was allocated for constructing the discharge rating curve, and the remaining 40% was dedicated to model validation.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Best Practice Approach to Stage-Discharge Rating Curve: Case Study of Selected Rivers in Selangor

Saadon,

Abdul Manaf,

Tholibon

et al. 2023

JSCET

View full text Add to dashboard Cite

A stage-discharge rating curve represents the relationship between the water level at a given point in a stream and a corresponding volumetric rate of flow. It is widely used for the analysis of flow regimes, designing hydraulic structures, flood forecasting, flow-sediment interaction and prediction, and river channel analysis. The reliability and stability of the stage-discharge rating curve are based on numerous numbers of discharges and stage data from the lowest to the highest bank full stage observed over a long period of time. Stage-discharge rating curves contribute to reliable discharge prediction for future flood forecasting. This study focuses on the best-practice approach to stage-discharge rating curve development for selected rivers in Selangor. The stagedischarge rating curve derived was evaluated in terms of the statistical parameter, which denotes the degree of determination or r-squared value. The polynomial function yields better accuracy at 68.6% than the exponential function at 60.3%. Model validation was performed using the discrepancy ratio and confirmed that the stagedischarge rating curve developed using both mathematical functions give good predictions with more than 80% accuracy for the specified flow range and stage levels.

show abstract

Section: Methodsmentioning

confidence: 99%

“…Values within the range of 0.5 to 2.0 are considered accurate. This evaluation method has been widely employed in previous studies, including those by Ibrahim et al (2017), Sinnakaudan et al (2010), Saadon et al (2021Saadon et al ( , 2020. The generated stage-discharge function with a higher D.R.…”

Section: Figure 1 Flow Chart Of the Overall Methodologymentioning

confidence: 99%

Best Practice Approach to Stage-Discharge Rating Curve: Case Study of Selected Rivers in Selangor

Saadon,

Abdul Manaf,

Tholibon

et al. 2023

JSCET

View full text Add to dashboard Cite

show abstract

“…Entre las ANN dinámicas destacan los modelos nonlinear autoregressive exogenous (NARX), que son considerados buenos predictores del comportamiento de sistemas dinámicos no lineales gracias a su rápida convergencia y su mejor capacidad de generalización, en comparación con otras redes (Manonmani et al, 2018). Las NARX's se han empleado en varias aplicaciones y han mostraron un rendimiento confiable (Saadon et al, 2020). La arquitectura de un modelo NARX se muestra en (3).…”

Section: Identificación Con Redes Neuronales Artificialesunclassified

Control de un sistema multivariable no lineal y en fase no mínima empleando un controlador predictivo neuronal

Chojeda

Semino

Ipanaqué

2022

Rev. iberoam. autom. inform. ind.

View full text Add to dashboard Cite

En este artículo se propone un Controlador Predictivo Neuronal (ANN-MPC) para controlar un sistema no lineal de tanque cuádruple, el cual es complejo de controlar debido a la no linealidad de sus válvulas y a la interacción entre sus variables controladas. Además, el problema se agrava ya que el proceso presenta una respuesta transitoria con dinámica inversa por estar en fase no mínima. El ANN-MPC emplea una estructura modular de red neuronal artificial y el algoritmo de entrenamiento Levenberg-Marquardt para estimar con mayor precisión y rapidez las salidas del proceso no lineal y evitar el sobreajuste del modelo. Se generaron datos operativos a partir de la planta para entrenar la red neuronal empleando Matlab. Se probó el rendimiento del ANN-MPC ante cambios de referencia y se comparó con un MPC lineal y un MPC no lineal. Los resultados de simulación mostraron que el ANN-MPC produjo un menor tiempo de establecimiento que el MPC lineal y generó valores RMSE de las salidas similares a los del NMPC. Además, se redujo el tiempo de cómputo requerido para calcular la variable de control óptima comparado con el NMPC.

show abstract

Application of Artificial Intelligence for Reservoir Storage Prediction: A Case Study

Azad

Vasant

Vintaned

et al. 2022

International Conference on Artificial Intelligence for Smart Community

View full text Add to dashboard Cite

Development of riverbank erosion rate predictor for natural channels using NARX-QR Factorization model: a case study of Sg. Bernam, Selangor, Malaysia

Cited by 12 publications

References 28 publications

Best Practice Approach to Stage-Discharge Rating Curve: Case Study of Selected Rivers in Selangor

Best Practice Approach to Stage-Discharge Rating Curve: Case Study of Selected Rivers in Selangor

Control de un sistema multivariable no lineal y en fase no mínima empleando un controlador predictivo neuronal

Application of Artificial Intelligence for Reservoir Storage Prediction: A Case Study

Contact Info

Product

Resources

About