Advanced search
Start date
Betweenand

Active control of DER-Rich distribution networks using artificial intelligence

Grant number: 19/25076-6
Support type:Scholarships abroad - Research Internship - Master's degree
Effective date (Start): May 09, 2020
Effective date (End): November 08, 2020
Field of knowledge:Engineering - Electrical Engineering - Power Systems
Principal researcher:Fernanda Caseño Trindade Arioli
Grantee:Guilherme de Oliveira Custodio
Supervisor abroad: Luis Fernando Ochoa Pizzali
Home Institution: Faculdade de Engenharia Elétrica e de Computação (FEEC). Universidade Estadual de Campinas (UNICAMP). Campinas , SP, Brazil
Research place: University of Melbourne, Australia  
Associated to the scholarship:18/23391-9 - Active control to increase pv hosting capacity of electrical power distribution network, BP.MS

Abstract

Electrical power distribution networks are likely to experience in the next decade significant infrastructural changes due to the massive integration of distributed energy resources (DER). As distribution networks become more observable and controllable, new modes on how to operate the system are available in such a way that the new assets, for instance, smart inverters of photovoltaic systems can contribute to the distribution network operator's planning and operation tasks. Ultimately, the coordination of all these controllable assets, both traditional (e.g. voltage regulators and capacitor banks) and modern devices, can make it possible to have large penetrations of DER in distribution networks without technical impacts. In this context, the main objective of this work is the investigation of the benefits and implementation challenges of an Active Control that harness DER control capabilities and of other controllable assets to help operate the system under regulatory policies. The development of the Active Control is proposed using innovative Machine Learning techniques that could eventually replace the need for detailed network models. Among the available techniques, supervised learning trained, for instance, with optimal power flow results, and reinforcement learning are going to be explored. The Active Control solutions will be compared and assessed according to their performance in mitigating voltage and thermal problems but also in terms of their practicality. (AU)

News published in Agência FAPESP Newsletter about the scholarship:
Articles published in other media outlets (0 total):
More itemsLess items
VEICULO: TITULO (DATA)
VEICULO: TITULO (DATA)