Work

Data Challenges and Data Analytics Solutions for Power Systems

Year: 2020
Type: dissertation
Abstract: The present work of thesis mainly presents some innovative data-driven applications to deal with the emerging data challenges in power system.With the overwhelming trend of digitalization in power system, critical challenges and opportunities are emerging in the modern industrial systems.In the first chapter, the big data analytics are introduced with corresponding applications in smart grids.With huge amount of data from electricity network, meteorological information system, geographical information system etc., many benefits can be brought to the existing network and import the customer service as well as the social welfare in the era of big data.In distribution network, there are a large number of feeders spread in a wide area with a diversity of structural features.Chapter 2 discusses the possibility to group thousands of feeders with a data-driven method and try to identify the representative feeders for a better evaluation of the performance with regard to the number of outages.Two clustering algorithms are applied on the mixed data with the similar analysis results.In Chapter 3, the data-driven solutions to the problems relating to the outage predictions in medium-voltage distribution network is presented.Annual and monthly number of outages are predicted based on improved grey theory.Further more, the daily outages prediction is discussed as a binary classification problem with support vector machine.Facing the challenges of limited channel bandwidth in advanced metering infrastructures, the consumption records with partial missing information are utilized in Chapter 4 to figure out the abnormal users in power system.By extracting the importance index through regression models, anomalies from the energy consumption patterns are successfully uncovered.First of all, I would like to thank my supervisor, professor Ettore Bompard.During the last few years, he accompanied and encouraged me to get out of the comfort zone and face new challenges to do a promising job.Thank you very much and I would forever be grateful for all his support.A special thank is for my parents, who supported me during my student career with unconditional love.A huge thank is for professor Andrea Mazza, who gave me a lot of guidance on the road of scientific research and always inspired me with innovative ideas.Furthermore, I would like to thank all my colleagues and friends: Shaghayegh and Francesco, who are enrolled in PoliTo at the same year with me and accompanied me the longest time; Renjian, my senior and also friend; Abouzar, who is always kind
Source: PORTO@Iris
Author: ZHANG, YANG
Language: English
Open Access status: green
OpenAlex ID: W7219289254
Cites: 165
Cited by: 1
Related to: 10