This paper introduces a new methodology to classify data extracted from off-grid solar installations, based on the charge and discharge processes. Using data from a real installation, we propose to classify the daily behaviour of the energy harvesting according to characteristics extracted from the voltage and current curves of the installation. Five type situations have been established, in which the off-grid system charge can be summarized. This classification can help to better understand the state of the installations and classify the behaviour of the charging process to train more complex algorithms.
Unsupervised clustering of battery waveforms in off-grid PV installations
2020-09-10
696944 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
AUTOMATED FACIES IDENTIFICATION USING UNSUPERVISED CLUSTERING
British Library Conference Proceedings | 2020
|Unsupervised Clustering of Highway Motion Patterns *
IEEE | 2019
|Image Segmentation by Unsupervised Sparse Clustering
IEEE | 2005
|Dignet: An Unsupervised-Learning Clustering Algorithm for Clustering and Data Fusion
Online Contents | 1995
|