Electrical and Computer Engineering ETDs
Publication Date
Spring 5-1-2020
Abstract
It has been 32 years since the Brundtland Report was published. That was the first time that the term Sustainable Development (SD) was coined. In this context, renewable sources of energy play an important role on not to deplete our natural resources in order to meet our need without compromising future generations. Smart Grids are on the pathway to achieve the SD goals. This Thesis focuses on the integration of renewables, specifically Solar PV panels and inverters and its interactions with the distribution grid. Since the power injection caused by the PV inverter can alter the voltage range, Reinforcement Learning (RL) is applied as method for voltage regulation. This research aims to integrate all these elements in a Co-simulation Real-Time system. To achieve complexity and reality to this cosimulation frame, an external load is aggregated from an external source. Methodology and results are described, and conclusions and future work suggested.
Keywords
reinforcement learning distribution feeder reactive power control simulation external aggregated load
Document Type
Thesis
Language
English
Degree Name
Electrical Engineering
Level of Degree
Masters
Department Name
Electrical and Computer Engineering
First Committee Member (Chair)
Manel Martinez-Ramon
Second Committee Member
Marios Pattichis
Third Committee Member
Ali Bidram
Recommended Citation
Acosta Molina, Ivonne D.. "INTEGRATION OF MACHINE LEARNING FOR REACTIVE POWER CONTROL FOR A DISTRIBUTION FEEDER SIMULATION WITH EXTERNAL LOAD." (2020). https://digitalrepository.unm.edu/ece_etds/520