A Methodology for Joint Data Cleaning of Solar Photovoltaic Generation and Solar Irradiation

A good solar power photovoltaic generation forecast depends on good quality time series data from measurements of global horizontal irradiation and solar generation. However, measurement system failures and errors in data handling can corrupt data records with gaps and outliers that undermine foreca...

Full description

Saved in:
Bibliographic Details
Main Authors: Pessanha, J. F. M., Melo, A. C. G., Caldas, R. P., Falcao, D. M.
Format: Conference Proceeding
Language:English
Subjects:
Data Cleaaning
Data models
Neurons
Predictive models
Radiation effects
Self-organizing feature maps
Self-Organizing Map
Solar power generation
Statistical Methods
Time Series
Training
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:A good solar power photovoltaic generation forecast depends on good quality time series data from measurements of global horizontal irradiation and solar generation. However, measurement system failures and errors in data handling can corrupt data records with gaps and outliers that undermine forecasting accuracy. Therefore, it is important that the fitting of solar energy prediction models must be preceded by a data analysis in order to detect and correct measurement errors. This paper presents the main features of an approach for the joint data cleaning of solar photovoltaic generation and solar irradiation. The methodology comprises 5 chained steps and consists in the combined treatment of global horizontal irradiation and solar photovoltaic generation data using statistical techniques, data mining algorithms and reanalysis data with the purpose of correcting outliers, replacing incorrect values and filling data gaps. The application of the proposed approach is illustrated with a real system presenting good performance.
ISSN:2642-6757
DOI:10.1109/PMAPS47429.2020.9183488