Title | ||
---|---|---|
Electricity Generation Forecasting in Concentrating Solar-Thermal Power Plants with Ensemble Learning |
Abstract | ||
---|---|---|
Electric power generation forecast systems in concentrating solar-thermal power plants are a key tool for their operation and maintenance optimization. The purpose of this work is to approach the problem of electric power prediction in Arenales concentrating solar-thermal plant (Sevilla, Spain). Throughout this work, the standard phases in the knowledge discovery in databases are followed, resulting in three different models for the hourly electric power forecasting, with a 24-h prediction horizon. Each model is based on a different algorithm: Extra Gradient Boosting, K-Nearest Neighbors, and a Multi-Layer Perceptron neural network. The fitness of the models is assessed by some of the most common error metrics in regression problems with a satisfactory result. Additionally, it is shown how the results obtained in the prediction of hourly energy give rise to also evaluate the daily and the aggregate energy prediction in a wide time interval. After an individual analysis of each model, a comparative study is included with the aim of determining the best performance model. |
Year | DOI | Venue |
---|---|---|
2021 | 10.1007/978-3-030-87869-6_63 | 16TH INTERNATIONAL CONFERENCE ON SOFT COMPUTING MODELS IN INDUSTRIAL AND ENVIRONMENTAL APPLICATIONS (SOCO 2021) |
Keywords | DocType | Volume |
Time series, Forecasting, Solar power, Renewable energy | Conference | 1401 |
ISSN | Citations | PageRank |
2194-5357 | 0 | 0.34 |
References | Authors | |
0 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
A. Melara | 1 | 0 | 0.34 |
J. F. Torres | 2 | 0 | 0.34 |
Alicia Troncoso | 3 | 153 | 20.88 |
Francisco Martínez-Álvarez | 4 | 0 | 0.34 |