Appropriate Operating Policy for a Reservoir System Based on Inflow States (Mosul Reservoir as a Case Study)

Abstract

The present study has focused on the selection of appropriate operating policies for a reservoir system to maximize the hydropower generation through the derivation of operation rule curves for different states of the inflow into the reservoir. And in order to accomplish this task, 30 years of monthly observed inflow has been used and arranged into five intervals with lower and upper values of each interval. The Genetic Algorithms techniques were implemented to obtain the optimal operating policies for each of the proposed five categories of inflow. Furthermore, and in order to alleviate the adverse effects of trapping in local maxima, a hybrid method is suggested making use of the built-in functions of optimization available in the Matlab software. The non-linear optimization function used the values of the variables obtained by GA's as initial solutions required by the optimization process to reach global solutions. One of the main objectives of the current study was to build a simple mechanism to select the most appropriate operating policy which is compatible with the nature of the inflow state entering the reservoir during a certain period of time. Consequently, a simulation model was conducted using SIMULINK technique to mimic the Mosul reservoir system which is selected as a case study to test the proposed method. The results reveal an improvement by 10 % of hydropower generation over that produced over the 30 years of real operation using the current method.