Dp 511 en
Development of algorithms for demand-side management in smart-grids
Author: Josef Müller
The aim of this thesis was to evaluate the possibility of using the existing algorithm in embedded devices based on ARM and make the necessary adjustments. The algorithm deals with optimizing the charging periods of water heaters in households in order to eliminate problems with renewable energy sources in low voltage areas.The first task was to get acquainted with the parameters and options of available development boards with ARM processor, compare these boards in terms of specified criteria and select the appropriate board for implementation of the control algorithm. This algorithm uses a two-phase optimization. The type of the optimization in the first phase is linear programming (LP) and in the second phase it is mixed integer linear programming (MILP). Therefore, the second task was to find for the selected development board a suitable non-commercial solver for these types of optimization problems. However, since the non-commercial solvers on the selected board could not find solution of MILP problems in a sufficiently short time, the final task of this thesis was to develop a new implementation of this part of the control algorithm. Thus was created a heuristic algorithm which solves the problem in a considerably shorter time and it provides a solution of similar quality, in some ways even better than the solution obtained by the original MILP optimization algorithm.