Data-Driven approach to topology change location in distribution networks using microPMUs

Motivated by the aim to increase the renewable energy penetration into the grid, the Mexican government established the objective of producing the half of its energy from clean sources by 2050. This is also a tendency in the rest of the world, but utilities are not yet prepared to deal with the challenges that the proliferation of this change will bring. A way to solve such issues is by evolving from the antiquated power system model to a smart grid, by building a control and communications infrastructure, and by introducing sensing and metering technologies. In this sense, micro-phasor measurement units (μPMU) are devices tailored for such purpose; but this technology requires specializing research in order to develop tools for its applications on field. Driven by this urgency, we established the objective of building an application based on the μPMU technology. Therefore, in this thesis we propose an algorithm to topology change location in distribution networks using μPMU data; based on a behavioral system theory in which we use any set of variables that are available for measurement within a network. Such approach differentiates from classic methods, since it does not require any information about the network model, and it does not assume any particular character of disturbance to locate the occurrence within the network. MATLAB simulations and experimentation using μPMUs and a DSpace Data Acquisition Card were implemented with satisfactory results, since the algorithm demonstrated to be capable to locate single topology changes in distribution networks.