MSc Thesis: Smart Reconfiguration of Electric Power Distribution Networks for Power Loss Minimisation and Voltage Profile Optimisation
Alberto Landeros Rojas successfully defends his master's thesis
REYKJAVIK, June 6 - As the 2017/18 academic year comes to an end, ISE master students finalise their thesis projects. Last month, Alberto Rojas, MSc in Sustainable Energy Engineering successfully defended his master's thesis project in smart reconfiguration of electric power distribution networks for power loss minimization and voltage profile optimisation. Alberto's work was supervised by Mohamed Abdelfattah and Slawomir Kozil, who are both from Reykjavik University.
According to Alberto, distribution network reconfiguration (DNR) significantly reduce power losses and increase power quality. In his project, Alberto introduced a customised evolutionary algorithm to power distribution network reconfiguration. The first method he applied was the feasibility-preserving evolutionary operator (FPEO), in which the 33 BUS, 69 BUS and 119 BUS were tested using population sizes of ten. The algorithm is designed to preserve the feasibility of solutions and hence, significantly reduce the size of the search space. In return, this improves the repeatability of results and at the same time, lower the overall computational complexity of the optimisation process. Alberto found that through this method, power losses were reduced by a quarter in the 33BUS network and after about 40 iterations, the configuration converged.
Next, the feasibility preserving simulated annealing (FPSA) method was further applied. This approach is adopted to solve DNR and is based on sequential stochastic optimization that uses mechanisms adopted from simulated annealing. The impetus behind this approach is to improve the cost function while maintaining the radial architecture of the distribution system. The FPSA method was tested under light, nomical and heavy-system loading. For all three scenarios, Alberto's method reports less power losses compared to other methods in the literature.
Alberto concluded that this work would hopefully not stop at his master thesis. Since switching costs are typically high, especially those in remote locations, future work could consider optimising switching locations. In addition, Alberto's thesis work has thus far, resulted in three scientific papers under peer review and four conference proceedings globally. To read more about Alberto's work, please click on the following link .
Congratulations, Alberto on an excellent thesis!