Date

2020

Author

Ercan, Eray

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

309
views

213
downloads

Cite This

In today's world, energy is one of the most important drivers for the continuation of civilization. Until recent years, the energy demands of the countries have been mainly supplied by fossil fuels. However, the negative effects of using fossil fuels in energy generation shifted the focus to renewable energy resources. In addition to this, the popularity of renewable energy resources increased as their costs decreased and efficiencies increased. Therefore, the integration of renewable energy systems to national grids increased in recent years. However, this integration is challenging due to the intermittent nature of renewable energy sources. In addition to the uncertainty in the generation of energy from the renewables, the uncertainty in the electricity spot markets increases the difficulty in the management of the renewables. To deal with the intermittent nature of the renewables, energy storage systems have to be implemented. Pumped storage hydropower is currently the most viable form of large scale energy storage. The operation of renewable systems, together with pumped storage hydropower plants, increases the efficiency of the hybrid system. In this study, a wind-hydro hybrid system (WHHS) is considered, and optimum daily operation strategies for a hypothetical case study is developed. To increase the revenue of the WHHS, a long short-term memory (LSTM) network is developed to forecast electricity prices in the day-ahead spot electricity market. Apart from the LSTM network, an optimization model is developed to obtain optimum operation schedules and the maximum revenue of the WHHS by using the electricity price and available wind energy as inputs. To investigate the effects of the LSTM network and the optimization model, different scenarios are created and run. According to the results, it is observed that wind turbines compensate the loss due to the poor forecasting of the electricity price. Thus, the higher the installed capacity of wind turbines in the WHHS, the better compensation it provides. However, within the studied range (i.e., 25 MW to 500 MW), the operation schedules of the pump and the hydro turbine of the WHHS are not affected from increasing the installed capacity of wind turbines. Once enough energy is generated by wind turbines to be used to pump the water to the upper reservoir, the rest of the wind energy is directly sold to the grid.

Subject Keywords

Renewable energy sources., Renewable Energy, Optimization, LSTM, Electricity Price Forecasting, Wind-Hydro Hybrid Systems.

URI

http://etd.lib.metu.edu.tr/upload/12625480/index.pdf
https://hdl.handle.net/11511/45700

Collections

Graduate School of Natural and Applied Sciences, Thesis