Development of an advection model for solar forecasting based on ground data first report: Development and verification of a fundamental model

History and trends in solar irradiance and PV power forecasting: A preliminary assessment and review using text mining

“…Abbreviation handling is one of the challenges in text mining, especially during the construction of lexical ontologies. Although most abbreviations are simply made of word initials, there are many cases for which the matching requires word skip (e.g., "randomized training and validation set method" is 1: Solar forecasting methods for renewable energy integration 2: Assessment of forecasting techniques for solar power production with no exogenous inputs 3: Intra−hour DNI forecasting based on cloud tracking image analysis 4: Review of solar irradiance forecasting methods and a proposition for small−scale insular grids 5: Solar radiation prediction using Artificial Neural Network techniques: A review 6: Solar energy forecasting and resource assessment 7: Evaluation of the WRF model solar irradiance forecasts in Andalusia (southern Spain) 8: Short−term solar irradiance forecasting model based on artificial neural network using statistical feature parameters 9: Weather modeling and forecasting of PV systems operation 10: A weather−based hybrid method for 1−day ahead hourly forecasting of PV power output 11: Hourly solar irradiance time series forecasting using cloud cover index 12: Short−term power forecasting system for photovoltaic plants 13: Numerical weather prediction (NWP) and hybrid ARMA/ANN model to predict global radiation 14: Short−mid−term solar power prediction by using artificial neural networks 15: Innovative second−generation wavelets construction with recurrent neural networks for solar radiation forecasting 16: Probabilistic energy forecasting: Global energy forecasting competition 2014 and beyond 17: Variability assessment and forecasting of renewables: A review for solar, wind, wave and tidal resources 18: A support vector machine−firefly algorithm−based model for global solar radiation prediction 19: Photovoltaic and solar power forecasting for smart grid energy management 20: An improved photovoltaic power forecasting model with the assistance of aerosol index data 21: Review of photovoltaic power forecasting 22: Developing a whole building cooling energy forecasting model for on−line operation optimization using proactive system identification 23: Analysis and validation of 24 hours ahead neural network forecasting of photovoltaic output power abbreviated as "RTM" by Chu et al, 2013), or using internal letters (e.g., "exponential smoothing" is abbreviated as "ETS" by Hyndman et al, 2008). Training-and learning-based abbreviation identification algorithms often fail, due to the everexpanding novel use of abbreviations.…”

“…Metrics such as RMSE, MAE and MBE are known as scaledependent errors (Hyndman et al, 2008). As such, they have limitations when comparing forecast accuracies across data with different scales.…”

“…A method is classified as a time series method if it falls in one of three families of models, namely, autoregressive integrated moving average (ARIMA), exponential smoothing (ETS), and generalized autoregressive conditional heteroskedasticity (GARCH). Largely owing to the two books written by Box and Jenkins (1994) and Hyndman et al (2008), the ARIMA and ETS families of models became iconic for the subject of statistical forecasting. The GARCH family of models-first proposed by Engle (1982) and then generalized by Bollerslev (1986)gained its recognition in econometrics due to its ability to perform heteroscedasticity corrections; it is widely used to model financial time series.…”

“…It has been applied several times in solar forecasting (e.g., Jamaly and Kleissl, 2017;Perez et al, 2016;Aryaputera et al, 2015b;Yang et al, 2014aYang et al, , 2013b. According to Cressie and Wikle (2011), however, another major method, namely, prediction using the stochastic partial differential equation (PDE), has not been well applied to solar sensor network data, at least until the recent publication by Inage (2017).…”

Fast Univariate Time Series Prediction of Solar Power for Real-Time Control of Energy Storage System

“…Multivariate solar prediction methods have already been significantly investigated in the literature [3][4][5][6][7][8][9]. In the Global Energy Forecasting Competition, 12 weather variables from the European Centre for Medium-range Weather Forecasts (ECMWF) were made available to the participants to generate probabilistic forecasts of three solar farms in Australia; the proposed methods are summarized in Ref.…”