2018 - PI World - Barcelona - Transmission & Distribution
Case Study: Short Term Demand Forecasting Methodology for Power Decision Making Based on Markov Chain
This investigation is focused on the prediction of the electrical demand in short term for the National Interconnected System of Ecuador. For this purpose, the proposed methodology trains a Gaussian Hidden Markov Model from historic data to determine the typical “demand profiles” of the electrical demand throughout the country. The process of discovering the demand profiles uses the historical data coming from a PI-Server and the hidden state sequence produced by the best trained HMM model.
Roberto Gonzalo Sánchez Albán
Roberto Sánchez Albán.- He was born in Quito in 1986. He received his Electronic Engineer degree from the University of the Armed Forces (Universidad de las fuerzas armadas ESPE) in 2010; of Master in Computer Science from the University of Bern/Neuchatel/Friborg - Switzerland in 2017. He worked as a SCADA engineer in the real-time area at the National Electricity Operator CENACE from 2009 to 2014. His fields of research are related to the application of Machine Learning, Artificial Intelligence and Data visualization in the industry, currently applied to the electrical sector. At the moment, he is working at Analysis of Operation of the electrical power network of Ecuador in the National Electricity Operator (CENACE).
Hector Patricio Barrera Navas
Patricio Barrera Navas.- He was born in Ambato in 1990. He received his degree in Electronic Engineering from the National Polytechnic School in 2016. He currently works at the National Electricity Operator, CENACE. His areas of interest lie in the post-operative analysis of power systems.