In this master’s thesis, we focus on the optimization of a non-linear program for the optimal operation of a battery storage system that is integrated with solar power plant. Aim of the optimization is to minimize the cost of electricity supply. optimizations are achieved using two approaches. Initially, we transform the program into a mixed integer linear program and solve it using the CBC algorithm. Then we solve the program using the interior point algorithm - Ipopt - while preserving the original non-convex form. The key concerns of this work are the effect of program transformation on the accuracy of the result and whether the optimization of the original form and the use of more time-consuming algorithms is reasonable. It has been shown that an algorithm CBC provides results with objective function values very close to the result obtained by solving the program in its original form, yet the optimization time is considerably shorter. However, although the objective function values are comparable, the two approaches yield different results, since the transformed form of the program tends to achieve results at the edge of the solution space, while the solutions of Ipopt remain within this range. Lastly, using the Tabu search algorithm, the neighborhood of both results is examined to confirm that there are no other results in the neighborhood that would result in a significantly improved objective function.
|
