Muhammad Ali Shamim

ANFIS and NNARX based rainfall-runoff modeling

Modeling of non-linearity and uncertainty associated with rainfall-runoff process has received a lot of attention in the past years. Recently artificial intelligence techniques are used for hydrological time series modelling. Earlier studies showed this approach is effective, still there are concerns about how these techniques perform efficiently to predict the run-off with high standard of accuracy. To this end, this paper explores the ability of two artificial intelligence techniques, namely neural network auto regressive with exogenous input (NNARX) and adaptive neuro-fuzzy inference system, to model the rainfall-runoff phenomenon effectively from antecedent rainfall and runoff information. Specifically, to illustrate applicability of these techniques, two year (1994-1995) rainfall-runoff data from Brue catchment of The United Kingdom were used. The models having various input structures were constructed and the best structure was investigated with help of the proposed technique, called gamma test. Training data length selection and best input combination were carried out prior to modeling with help of gamma test. The performance of the ANFIS model in training and testing sets were compared with that of NNARX model with help of several statistical parameters. The results of the study have shown that both ANFIS and NNARX could work efficiently in rainfall-runoff modeling and can provide high accuracy and reliability in runoff prediction.

A hybrid modelling approach for assessing solar radiation

A hybrid technique for solar radiation estimation, a core part of hydrological cycle, is presented in this study which parameterises the cloud cover effect (cloud cover index) not just from the geostationary satellites but also the PSU/NCAR’s Mesoscale Modelling system (MM5) model. This, together with output from a global clear sky radiation model and observed datasets of temperature and precipitation are used as inputs within the Gamma test (GT) environment for the development of nonlinear models for global solar radiation estimation. The study also explores the ability of Gamma test to determine the optimum input combination and data length selection. Artificial neural network- and local linear regression-based nonlinear techniques are used to test the proposed methodology, and the results have shown a high degree of correlation between the observed and estimated values. It is believed that this study will initiate further exploration of GT for improving informed data and model selection.

Predicting streamflows to a multipurpose reservoir using artificial neural networks and regression techniques

Population increase and climate change are stretching not only the world’s but also Pakistan’s water resources. This has directly been responsible for the recurring patterns of floods and droughts in the country which emphasizes the importance of the fact that efficient practices need to be adopted for water resource sustainability. This study investigates the use of upland catchment information, comprising of hydrometeorological datasets for inflow prediction to the Tarbela reservoir (a multipurpose reservoir located on River Indus) using Artificial Neural Networks (ANN) and Regression Techniques (Standard and Step Wise). Input Combination and data length selection for all the selected techniques were performed with the aid of Gamma test (GT). This study has made a significant contribution for future water resource management within the Indus Basin as Tarbela is the main source of irrigation, water supply and hydropower generation in Pakistan along with flood control.

EFFICIENT TREATMENT OF DAIRY EFFLUENT CONTAINING RICH NUTRIENTS USING UP-FLOW MULTILAYER BIO-REACTOR (UMBR)

Two pilot-scale biological nutrient removal systems having combination of up-flow multilayer bio-rector (UMBR), aeration tank (AT) and sedimentation tank (ST) were installed to check their treatment efficiency for replicated and dairy effluent. These systems were operated under five stages to achieve better results. Hydraulic retention time (HRT), inner recycle (IR), sludge recycle (SR) and organic loading rate (OLR) were systematically changed to optimize the operational conditions. System-I and system-II were operated for dairy and replicated effluent respectively for appropriate analysis. De-nitrification efficiency in up-flow multilayer bioreactor (UMBR) was observed significantly during the operation. It was observed that the involvement of UMBR improved the overall Total Chemical Oxygen Demand (TCOD), Soluble Chemical Oxygen Demand (SCOD). Nitrite-Nitrogen (NO3-N). Total Kjeldahl Nitrogen (TKN), Nitrate Nitrogen (NH3-N) and Total Phosphorous (TP) removal efficiency as compared to other traditional biological treatment systems. Both systems showed practical treatment of replicated as well as dairy effluent. These systems were operated under different set of operational mode and collected results are discussed in term of treatment efficiency of different components. The overall treatment efficiency of the combined systems was also discussed.

FLOOD ESTIMATION BY VARIOUS TECHNIQUES FOR SMALL AND LARGE CATCHMENTS

Rainfall runoff relation is very complex and depends on so many factors which require numerous assumptions. Misleading results may be obtained due to over simplification of the process. The various formulas developed for estimation of peak flood discharge are generally based on single geometric parameter, the catchment area which does not reflect true discharges on which hydraulic structures like protection gabion walls, weirs, protection embankments, small irrigation weirs built on torrential flood channels, culverts and bridges are designed. This results in loss of precious money of the state and repeated rehabilitations. This paper presents a procedure for estimating the peak flood discharge. A novel regional mathematical rainfall-runoff model was developed by full convolution of the synthetic unit hydrograph known as Snyder Method extended to include design return period of the selected storm and aerial reduction factor analysis. The model was tested against actual stream flow records by incorporating the flow records in the form of a regional flood frequency curve. The Synthetic Unit hydrograph and the design storm are related to find flood hydrograph corresponding to the design storm which may be routed through the hydraulic structure. Finally the developed mathematical model was analyzed for variation with respect to parameters of which it is composed of. The results were found within satisfactory limits of statistical tests.