Year: 2024
Type: conference-paper
Abstract: This paper investigates the novel performance analysis of photovoltaic (PV) installations by applying machine learning techniques (ML). The data used for the research is a mixed dataset of data from an experimental PV installation located at Brunel University London with correspondingly available weather data. Firstly, the analysis aims to establish various sensitivity relationships between PV power generation and weather conditions using techniques based on Random Forest ML. Secondly, the processing stage is implemented to assess the fitness of the different ML techniques through cross-validation. The results highlight the differing effectiveness of the applied approaches in achieving accurate and reliable results for the PV installations. The best techniques offer valuable insights for optimizing renewable energy usage in diverse environmental conditions.
Authors: Daniil Hulak, Yang Xie, Gareth Taylor
Institutions: Brunel University of London, Southeast University
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W4407938363