Student:	Ahsen ULUTAŞ
Supervisor:	Prof. Dr. İsmail Hakkı ALTAŞ
Department:	Elektrik-Elektronik Müh.
Institution:	Graduate School of Natural and Applied Sciences
University:	Karadeniz Technical University Turkey

Title of the Thesis:	NEURO-FUZZY BASED SMART ENERGY MANAGAMENT IN MICROGRIDS
Level:	M.Sc.
Acceptance Date:	20/6/2019
Number of Pages:	81
Registration Number:	i3614

Summary:

The share of renewable energy sources in energy generation has been increased

termandously during last decades. Hybrid systems have been created by combining more

than one renewable resources in terms of energy sustainability and efficiency. Storage units

have been added to enhance the stability of the hybrid systems. Storage units have made the

energy more feasible especially during peak hours. In these hybrid systems, energy

management is carried out for uninterrupted and stable operation.

The aim of this thesis is to provide a day ahead prediction of the energy to be generated

from renewable sources before the day and provide an intelligent energy management

system based on Neuro-Fuzzy to meet the energy demand of the consumers and ensure the

stable operation of the network. In this thesis, the possible working conditions of a hybrid

energy system consisting of FV, wave and battery are determined and a smart energy

management algorithm based on neuro-fuzzy has been realized in order to ensure stable

operation of the system under various conditions.

The day ahead energy prediction is done by estimating solar radiation and sea

wavelength using an artificial neural network method. Then the intelligent energy

management is done by a Neuro-fuzzy based algorithm. The grid-connected hybrid system

is modelled in the MATLAB/simulink simulation program. The simulation results show the

applicability of the proposed of the day ahead energy prediction and intelligent management

system.