Mehmet Ardiclioglu

Hydropower Optimization for the Lower Seyhan System in Turkey using Dynamic Programming

Abstract Dynamic programming with successive approximation has been used in the past for optimizing multi-reservoir water resources systems. In this study, the State Incremental Dynamic Programming (SIDP) model is developed for energy optimization of multi-reservoir systems. A random file access method is used for reaching initial and intermediate data to cope with the curse of dimensionality of dynamic programming. A conventional dynamic programming method is used for each single reservoir to find the initial trajectory of the reservoirs. Then, the computer program developed in the study is applied to the multipurpose-multi-reservoir system in Lower Seyhan Basin, which has six reservoirs, some of which are serial and some parallel. First, extended historical flows were used to maximize firm energy in the critical period, and then total energy in the total flows. The program was run with 50-year long segments (20 flow scenarios) of the synthetic flow data generated by using the HEC-4 generalized computer program to take into account the stochastic nature of stream flows. An increment of approximately 20 percent in total energy was obtained by using the model for the Lower Seyhan System, as compared to that calculated previously by conventional methods.

An artificial neural network model for the prediction of critical submergence for intake in a stratified fluid medium

Density differences may occur because of temperature differentials, suspended sediments, dissolved salts or other chemicals. Most of the large surface reservoirs are stably stratified throughout most, or all, of the year. One of the means of assisting the management is to allow a selective withdrawal from the reservoir. And while an intake is used for withdrawal (from the lower layer), a maximum discharge is required not allowing the uptake of the upper layer fluids. The value of the intakes vertical distance from the upper layer elevation (submergence) when the upper layer fluids begin to be drawn into the intake is known as ‘critical submergence’. In this study, the critical submergence for a circular intake pipe in a stratified body (which has different layer thickness) is investigated. Experiments were conducted on a vertically flowing downward intake pipe in a still-water reservoir. Artificial neural network (ANN) models and formulas, which are found by the theoretical analysis of critical spherical sink surface (CSSS), are used for the analysis of experimental results. The CSSS has the same centre and discharge as the intake. The ANN model and CSSS results are compared with the experimental results.

A Novel Study for the Modeling of Monthly Evaporation Using K-Nearest Neighbor Algorithms for a Semi-Arid Continental Climate

This study aims to reveal a reliable and efficient method for predicting the monthly evaporation. For this purpose, the accuracy of machine learning algorithms, MLA, that include k-nearest neighbor, k-NN, was used in modeling monthly evaporation. The tenfold cross-validation approach was employed to determine the performances of prediction methods for MLA. The results revealed that k-NN algorithms outperformed the other MLA (ANN and SVM), with the R value of 0.95, the RMSE value of 1.01 mm, MAE value of 0.78 mm, and RME value of 0.04 mm. It is concluded that the suggested k-NN model can be successfully employed for predicting monthly evaporation for a semi-arid continental climate.