Institutional Repository of Vilnius University
Fault detection in the solar power generation process /
Abstract
dc:descriptionIn this paper, statistical and machine learning methods are employed to detect faults in the solar power generation process. The primary objective is to analyze unlabeled data and identify strings affected by shading. The behavior of 143 strings that generate solar electrical power is investigated, comparing them based on their similarities. Initially, the data is evaluated using mean, standard deviation, and correlation coefficients to detect any abnormalities. Following this, K-means clustering is applied to the data from the days with the highest power generation and compared with the results with those from average generation days. Next, the time series data is transformed into functional data and smoothed using Fourier basis functions. This functional data is then used to detect outliers and perform functional K-means clustering.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2024
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Galinytė, Monika,
Subjects
dc:subject × 1Rights
dc:rights- Statement dc:rights
-
- info:eu-repo/semantics/openAccess
- Language dc:language
- eng
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://repository.vu.lt/VU:ELABAETD204759815&prefLang=en_US