The study aims to predict Total Dissolved Solids (TDS) as a water quality indicator parameter at spatial and temporal distribution of the Tigris River, Iraq by using Artificial Neural Network (ANN) model. This study was conducted on this river between Mosul and Amarah in Iraq on five positions stretching along the river for the period from 2001to 2011. In the ANNs model calibration, a computer program of multiple linear regressions is used to obtain a set of coefficient for a linear model. The input parameters of the ANNs model were the discharge of the Tigris River, the year, the month and the distance of the sampling stations from upstream of the river. The sensitivity analysis indicated that the distance and discharge have the most significant affect on the predicted TDS concentrations. The results showed that a network with (8) hidden neurons was highly accurate in predicting TDS concentration. The correlation coefficient (r), root mean square error (RMSE) and mean absolute percentage error (MAPE) between measured data and model outputs were calculated as 0.975, 113.9 and 11.51%, respectively for testing data sets. Comparisons between final results of ANNs and multiple linear regressions (MLR) showed that the ANNs model could be successfully applied and provides high accuracy to predict TDS concentrations as a water quality parameter.
There is a scarcity of data regarding algal flora of Tigris River in the territory of Baghdad. The present study deals with Tigris River in Al-Dora site in Baghdad province from November 2014 to June 2015 in order to shed light on its epiphytic Algae on (Phragmites australis) and epipelic algae. An amount of 183 and 154 species of epiphytic and epipelic algae are identified respectfully. The Bacillariophyceae (diatoms) are the dominant algal group followed by Cyanophyceae and Chlorophyceae. Moreover, 90 species are shared between two groups of algae (epiphytic and epipelic) and identified at the study site. Additionally, the seasonal variations and diversity of algal species are noticed. The highest number of epiphytic algae is 772.05 x 104
... Show More
(2)
In this paper we describe several different training algorithms for feed forward neural networks(FFNN). In all of these algorithms we use the gradient of the performance function, energy function, to determine how to adjust the weights such that the performance function is minimized, where the back propagation algorithm has been used to increase the speed of training. The above algorithms have a variety of different computation and thus different type of form of search direction and storage requirements, however non of the above algorithms has a global properties which suited to all problems.
(31)
(21)
This study aims to predict the organic pollution produced from the presence of some polycyclic aromatic hydrocarbons (PAHs) and determination it's concentrations (µg/L , ppb) in Tigris river water by a collection twenty-seven water samples from a selected three stations with nine sampling sites and three depths of water (5 cm , 2 m and 4 m) each site for 4.6 km distance of a geographic studied area which is located between the ( Al-Senak and AL-Sarrafiah bridges ) at Baghdad city – Iraq on May, 2012. The geographic location was determined with a Global Positioning System (GPS) and Geographic Information System (GIS) software program. The concentrations of fourteen components (PAHs) were performed using the reverse phase
... Show More
(16)
It is well known that drilling fluid is a key parameter for optimizing drilling operations, cleaning the hole, and managing the rig hydraulics and margins of surge and swab pressures. Although the experimental works represent valid and reliable results, they are expensive and time consuming. In contrast, continuous and regular determination of the rheological fluid properties can perform its essential functions during good construction. The aim of this study is to develop empirical models to estimate the drilling mud rheological properties of water-based fluids with less need for lab measurements. This study provides two predictive techniques, multiple regression analysis and artificial neural networks, to determine the rheological
... Show More
(30)
The present study conducted to study epipelic algae in the Tigris River within Baghdad city for one year from September 2011 to August 2012 due to the importance role of benthic algae in lotic ecosystems. Five sites have been chosen along the river. A total of 154 species of epipelic algae was recorded belongs to 45 genera, where Bacillariophyceae (Diatoms) was the dominant groups followed by Cyanophyceae and Chlorophyceae. The numbers of common types in three sites were 47 species. Bacillariophyceae accounted 88.31% of the total number of epipelic algae, followed by Cyanophyceae 7.14 % and Chlorophyceae 4.55%. A 85 species (29 genera) recorded in site 1, 103 species (34 genera) in site2, 112 species (35 genera) in site3, 96 species
... Show More