Akıllı mikro-şebekelerde kontrol stratejilerinin geliştirilmesi

This thesis concerns the transformation of aged power systems to modern power systems that include microgrids with renewable energy sources and energy storage systems. The integration of renewable energy sources brings excellent opportunities to provide better reliability and efficiency. However, the uncertainty and intermittent nature of renewable energy sources may potentially degrade the stability and quality of the electrical grid. Therefore, the aim of this dissertation is to maintain the supply-demand balance in microgrids while minimizing the cost in real time operation. A microgrid energy management system that can optimize the dispatch of the controllable distributed energy resources in grid-connected mode of a pilot microgrid on a university campus in Malta was developed to achieve this goal. Three different methods were used in this study: mixed integer linear programming (MILP), MILP based rolling horizon control and Q-learning, Designing intelligent method for the real-time energy management of the stochastic and dynamic microgrid is the primary goal of this research. Moreover, the detailed mathematical models of the network model and of the technical model are considered for the economic and environmental operation of the microgrid system to solve the optimization problem under more real-world conditions. The objective is to minimize the total daily operation costs, which include the degradation cost of batteries, the cost of energy bought from the main grid, the fuel cost of the diesel generator, and the emission cost. The optimization problem is modeled as a finite Markov decision process (MDP) by combining network and technical constraints, and a Q-learning algorithm is adopted to solve the sequential decision subproblems. The proposed algorithm decomposes a multi-stage mixed-integer nonlinear programming (MINLP) problem into a series of single-stage problems so that each subproblem can be solved using Bellman's equation. To prove the effectiveness of the proposed algorithm, three different case studies are taken into consideration. A predictive control framework is also proposed to provide optimal operation with minimum cost. This method allows the consideration of operational cost values, demand with uncertainty, generation units' profiles with uncertainty, and constraints related to the network model and technical model. The stochastic and deterministic cases are conducted to validate the efficiency of the approach.