Enhancing the Performance of Renewable Energy Systems via Energy Storage Capabilities

Abstract

In this work, one of the renewable energy resources, wind systems, has been selected for performance enhancement. There are many pros of the wind energy systems such as they don't produce harmful gas emissions, don't require fossil fuel, and can be installed in remote areas. However, these systems have a fluctuating output power. This power fluctuations problem is considered an important problem because of the bad impacts on the system performance. Thus, solving this problem becomes important to get an improved performance. A method is presented in this thesis which is used for reducing the power fluctuations output from any wind energy system connected to the utility grid. The proposed method depends on using the storage battery, the diesel generator and the dump load via three scenarios. In the first scenario, the storage battery operates only with the dump load. In the second scenario, the storage battery and the diesel generator operate together with the aid of the dump load. In the third scenario, the diesel generator operates only with the dump load. The economic, the environmental, and the technical aspects are taken into consideration. The thesis’s main objective is enhancing the wind energy system’s performance. This is done by solving the technical and the environmental problems and maintaining a high profit from selling the energy to the grid. An optimization problem with the proper constraints is formulated for the three scenarios. The Firefly Algorithm (FA) technique is used for finding these optimization problems’ solutions. The Harmony Search Algorithm (HSA) technique is also used for checking the results which are obtained by the FA and hence a comparison is done between them. The obtained results and graphs show that the proposed mechanism improve the wind energy system’s performance by reducing the power fluctuations within the acceptable limits and gaining a profit. The second scenario is considered the best used scenario in this proposed mechanism. Both techniques are able to find very close solutions for all tested number of populations and iterations. However, the FA produces slightly better results than the HSA.