Using Combinational Methods for Forecast Improvement in PV Power Plants


Abstract:

Power generation based on photovoltaic systems are one of the crucial energy resources of the future smart grid owing to environmental concerns. With the increasing involvement of photovoltaics into the smart grid, more accurate and reliable generation forecasts are essential for robust grid operations. In this paper, the authors have used the combining forecast method with forecasts generated from five different techniques, two of which use physical forecasting models and three of them use time series models. The methods are applied to simulation data of 12 PV power plants in the California region. Applying the combinational method to the generated forecasts, 25%33% accuracy improvements are observed compared to the reference model.

Año de publicación:

2018

Keywords:

    Fuente:

    scopusscopus

    Tipo de documento:

    Conference Object

    Estado:

    Acceso restringido

    Áreas de conocimiento:

    • Fotovoltaica
    • Aprendizaje automático
    • Energía renovable

    Áreas temáticas:

    • Física aplicada
    • Otras ramas de la ingeniería