OPTIMIZATION OF SIAH BISHE PUMPED STORAGE POWER PLANT DAILY SCHEDULE USING A GENETIC ALGORITHM MODELING FRAMEWORK

Energy storage is an issue that scientists have been researching for many years. There are several energy storage alternatives, among which the pumped storage system, battery storage, superconducting magnets, ywheels, and compressed air are the most popular. Based on former research, pumped storage is the best method for energy storage. The pumped storage plant is an effective method to balance the dierence value between electricity demand and production, which has greatly expanded. The advantages of this system include a considerable capacity to store energy, a short-term response to power network uctuations, long-term energy storage and etc. Although there are many pumped storage plants under service, a comprehensive research have yet to be undertaken. The purpose of this research is optimization of daily planning, based on available data of an existing pumped storage plant. To achieve this aim, minimization of the dierence value between energy production and demand has been determined. A genetic algorithm was used for optimization. The genetic algorithm is a meta-heuristic algorithm. It is inspired by the genetic evolution of living creatures. This algorithm was used according to binary coding. The Siah Bishe pumped storage plant in Iran was the research case study. This plant includes two non-leveled reservoirs that are connected by two water tunnels. The dierence level of the reservoirs is approximately 500 meters. Two reversible pump-turbine units are applied in each tunnel, and all four pump-turbines have the same specications. There are three constraints for this problem: 1) the maximum and minimum water volume of upper and lower reservoirs, 2) the maximum and minimum water ow of the pump-turbine unit and 3) the stationary of exploitation. Based on the international electricity consumption pattern, it is assumed that energy consumption is constant during one hour, while it may change each individual hour. Finally, the solution is estimation of the ow rate into the tunnels over 24 hours.