login

Search documents

Browse topics

Document details

 
Title:
 
Automated Performance Monitoring of Multiple Rooftop Systems Using a Single Machine Learning Algorithm
 
Author(s):
 
K. Shetty, Y. Kaushal, R. Dhavan, V. Murthy
 
Keywords:
 
Data Mining, Performance Monitoring, Machine Learning, Rooftop PV Systems, Python
 
Topic:
 
PV Systems and Storage – Modelling, Design, Operation and Performance
Subtopic: Operation, Performance and Maintenance of PV Systems
Event: 36th European Photovoltaic Solar Energy Conference and Exhibition
Session: 5BO.5.5
 
Pages:
 
1288 - 1291
ISBN: 3-936338-60-4
Paper DOI: 10.4229/EUPVSEC20192019-5BO.5.5
 
Price:
 
 
0,00 EUR
 
Document(s): paper
 

Abstract/Summary:


The aim of this work is to develop a machine-learning algorithm, which will have the capability to automatically fetch the performance data of several rooftop systems across a chosen region of interest on a periodic basis. The performance data of every rooftop site will be analysed by comparison to other sites of same installed capacity and compared against the estimated power generation of each site. Estimation data is normally calculated based on site’s geographical location, latitude, solar irradiance, plant capacity, installation angle of PV modules and other site related aspects. The machine-learning model created with such inputs will be able to generate an output that shows site-to-site and inverter-to-inverter level performance comparison in a graphical manner through a data visualization platform.