Online short-term solar power forecasting

By P. Bacher, Henrik Madsen, Henrik Aa. Nielsen. 2009

816 citations recorded by the source corpus.

This paper describes a new approach to online forecasting of power production from PV systems. The method is suited to online forecasting in many applications and in this paper it is used to predict hourly values of solar power for horizons of up to 36 h. The data used is 15-min observations of solar power from 21 PV systems located on rooftops in a small village in Denmark. The suggested method is a two-stage method where first a statistical normalization of the solar power is obtained using a clear sky model. The clear sky model is found using statistical smoothing techniques. Then forecasts of the normalized solar power are calculated using adaptive linear time series models. Both autoregressive (AR) and AR with exogenous input (ARX) models are evaluated, where the latter takes numerical weather predictions (NWPs) as input. The results indicate that for forecasts up to 2 h ahead the most important input is the available observations of solar power, while for longer horizons NWPs are the most important input. A root mean square error improvement of around 35% is achieved by the ARX model compared to a proposed reference model.

Subfields: Artificial Intelligence, Electrical and Electronic Engineering, Renewable Energy, Sustainability and the Environment

View Paper: https://www.semanticscholar.org/paper/0d2cd70c135a15a6d7c73675404cacaade435ade

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Electrical and Electronic Engineering

In Renewable Energy, Sustainability and the Environment

Citations in the corpus

Accurate photovoltaic power forecasting models using deep LSTM-RNN (2017, methodology): “Bacher et al. [2] proposed a two-stage method that incorporates auto-regressive model and auto-regressive with exogenous input model for short-term forecasting of PV power.”
Very‐short‐term probabilistic forecasting of wind power with generalized logit–normal distributions (2012, background): “…(Tay and Wallis, 2000; Timmermann, 2000), earthquake prediction (Kagan and Jackson, 2000) and meteorology (Leutbecher and Palmer, 2008), as well as for weather-related processes like renewable energy production (Bacher et al., 2009; Pinson et al., 2007) or floods (Cloke and Pappenberger, 2009).”
WRF-Solar: Description and Clear-Sky Assessment of an Augmented NWP Model for Solar Power Prediction (2016, background): “Regardless of the approach selected, the challenge of solar power forecasting has been recognized to be essentially the same as for solar irradiance forecasting (Bacher et al. 2009).”
Photovoltaic and solar power forecasting for smart grid energy management (2015, methodology): “and taking NWPs as input is developed to predict hourly solar power with look-ahead time up to 36 h, and tested on 21 PV systems located in a small village in Denmark [14].”
Forecasting of global and direct solar irradiance using stochastic learning methods, ground experiments and the NWS database (2011, background): “…here compare favorably to those of satellite-based models (Perez et al., 2002; Schillings et al., 2004; Vignola et al., 2007), and other medium-range (lead times of 1–6- days) forecasting models (Lorenz et al., 2009; Bacher et al., 2009; Breitkreuz et al., 2009; Perez et al., 2007, 2010).”
Generation and evaluation of space–time trajectories of photovoltaic power (2016, methodology): “Therefore, inspired by [34], quantile regression as a statistical smoothing method is deployed here.”
Solar Power Prediction Using Interval Type-2 TSK Modeling (2013, methodology): “We compare the performance of the three models to models of the same plant obtained using the methods of [8] and [9].”
Discrimination ability of the Energy score (2013, background): “This phenomenon touches a wide range of applications, from economics and finance [1, 2], to earthquake prediction [3], while it also has wide appeal in meteorology [4], and for weather-related processes like renewable energy production [5, 6] and floods [7].”
Review of photovoltaic power forecasting (2016, methodology): “Here we find Auto-Regressive (AR) models (Bacher et al., 2009), which model the output as a linear combination of lagged values of the predictors; simple Moving Average (MA) models (Li et al., 2014), used when data presents a constant variance over an equilibrium position around the mean, for which…”
An analog ensemble for short-term probabilistic solar power forecast (2015, background): “In [19] it is shown that for forecasts up to 2 h ahead the most important input is the available observations of solar power, while for longer horizons numerical weather prediction (NWP) model output becomes crucial for a better accuracy.”