Hydroelectric Flow Optimization of a Dam: A Kriging-Based Approach

Abstract

This paper presents a novel approach for hydroelectric flow optimization. The proposed approach integrates Kriging into the framework of genetic algorithm (GA). This coupling reduces the computational effort associated with conventional GA without affecting the accuracy. The proposed approach has been used for hydroelectric flow optimization. For the model considered, revenue generated is dependent on (a) hourly turbine flow, (b) hourly spill flow, (c) hourly electricity price and (d) storage level of reservoir. Two case studies have been performed by varying the simulation time considered. For the first case, the simulation is run for 50 h. It is observed that the proposed approach yields accurate result at significantly reduced computational cost. On contrary for the second case, the simulation is run for 20 days. Due to huge computational cost involved, it was not possible to generate benchmark solution. Hence, only the results obtained using the proposed approach have been reported. The results obtained are indicative of the fact that the proposed approach can be utilized for optimization of large-scale system from an affordable computational cost.