An ARMAX model for forecasting the power output of a grid connected photovoltaic system

Power forecasting has received a great deal of attention due to its importance for planning the operations of photovoltaic (PV) system. Compared to other forecasting techniques, the ARIMA time series model does not require the meteorological forecast of solar irradiance that is often complicated. Due to its simplicity, the ARIMA model has been widely discussed as a statistical model for forecasting power output from a PV system. However, the ARIMA model is a data-driven model that cannot take the climatic information into account. Intuitively, such information is valuable for improving the forecast accuracy. Motivated by this, this paper suggests a generalized model, the ARMAX model, to allow for exogenous inputs for forecasting power output. The suggested model takes temperature, precipitation amount, insolation duration, and humidity that can be easily accessed from the local observatory as exogenous inputs. As the ARMAX model does not rely forecast on solar irradiance, it maintains simplicity as the conventional ARIMA model. On the other hand, it is more general and flexible for practical use than the ARIMA model. It is shown that the ARMAX model greatly improves the forecast accuracy of power output over the ARIMA model. The results were validated based on a grid-connected 2.1 kW PV system.

[1]  Lennart Ljung,et al.  System Identification: Theory for the User , 1987 .

[2]  Mohamed Mohandes,et al.  Estimation of global solar radiation using artificial neural networks , 1998 .

[3]  M. A. Wincek.  Forecasting With Dynamic Regression Models , 1993 .

[4]  A. K. M. Sadrul Islam,et al.  Potential and viability of grid-connected solar PV system in Bangladesh , 2011 .

[5]  R. Belmans,et al.  Voltage fluctuations on distribution level introduced by photovoltaic systems , 2006, IEEE Transactions on Energy Conversion.

[6]  Ajeet Rohatgi,et al.  Design Considerations for Large Roof-integrated Photovoltaic Arrays , 1997 .

[7]  Richard A. Davis,et al.  Time Series: Theory and Methods , 2013 .

[8]  Shuanghua Cao,et al.  Forecast of solar irradiance using recurrent neural networks combined with wavelet analysis , 2005 .

[9]  Henrik Madsen,et al.  Online short-term solar power forecasting , 2009 .

[10]  William W. S. Wei,et al.  Time series analysis - univariate and multivariate methods , 1989 .

[11]  S. Safi,et al.  Modelling solar half-hour data using fourth order cumulants , 2002 .

[12]  Lei Wang,et al.  An ANN-based Approach for Forecasting the Power Output of Photovoltaic System , 2011 .

[13]  Tony N.T. Lam,et al.  Artificial neural networks for energy analysis of office buildings with daylighting , 2010 .

[14]  Volker Coors,et al.  Large scale integration of photovoltaics in cities , 2012 .

[15]  K. Sopian,et al.  Predicting average energy conversion of photovoltaic system in Malaysia using a simplified method , 2004 .

[16]  F. Hocaoglu,et al.  Hourly solar radiation forecasting using optimal coefficient 2-D linear filters and feed-forward neural networks , 2008 .

[17]  L. Zarzalejo,et al.  Prediction of global solar irradiance based on time series analysis: Application to solar thermal power plants energy production planning , 2010 .

[18]  Llanos Mora-López,et al.  Machine Learning Approach for Next Day Energy Production Forecasting in Grid Connected Photovoltaic Plants , 2011 .

[19]  Montserrat Mendoza-Villena,et al.  Short-term power forecasting system for photovoltaic plants , 2012 .

[20]  George E. P. Box,et al.  Time Series Analysis: Forecasting and Control , 1977 .

[21]  Cyril Voyant,et al.  Forecasting of preprocessed daily solar radiation time series using neural networks , 2010 .

[22]  S. Safi,et al.  Prediction of global daily solar radiation using higher order statistics , 2002 .

[23]  Gordon Reikard.  Predicting solar radiation at high resolutions: A comparison of time series forecasts , 2009 .

[24]  Ismail Musirin,et al.  Performance Analysis of Evolutionary ANN for Output Prediction of a Grid-Connected Photovoltaic System , 2009 .

[25]  Tom E. Baldock,et al.  Feasibility analysis of renewable energy supply options for a grid-connected large hotel , 2009 .

[26]  C. Holt.  Author's retrospective on ‘Forecasting seasonals and trends by exponentially weighted moving averages’ , 2004 .

[27]  Bangyin Liu,et al.  Online 24-h solar power forecasting based on weather type classification using artificial neural network , 2011 .

[28]  H. Pedro,et al.  Assessment of forecasting techniques for solar power production with no exogenous inputs , 2012 .

[29]  Jorge Aguilera,et al.  Generation of hourly irradiation synthetic series using the neural network multilayer perceptron , 2002 .

[30]  Peter R. Winters,et al.  Forecasting Sales by Exponentially Weighted Moving Averages , 1960 .

[31]  Athanasios Sfetsos,et al.  Univariate and multivariate forecasting of hourly solar radiation with artificial intelligence techniques , 2000 .

[32]  G. Schwarz.  Estimating the Dimension of a Model , 1978 .

[33]  A. Mellit,et al.  A 24-h forecast of solar irradiance using artificial neural network: Application for performance prediction of a grid-connected PV plant at Trieste, Italy , 2010 .