Forecasting the daily power output of a grid-connected photovoltaic system based on multivariate adaptive regression splines

Both linear and nonlinear models have been proposed for forecasting the power output of photovoltaic systems. Linear models are simple to implement but less flexible. Due to the stochastic nature of the power output of PV systems, nonlinear models tend to provide better forecast than linear models. Motivated by this, this paper suggests a fairly simple nonlinear regression model known as multivariate adaptive regression splines (MARS), as an alternative to forecasting of solar power output. The MARS model is a data-driven modeling approach without any assumption about the relationship between the power output and predictors. It maintains simplicity of the classical multiple linear regression (MLR) model while possessing the capability of handling nonlinearity. It is simpler in format than other nonlinear models such as ANN, k-nearest neighbors (KNN), classification and regression tree (CART), and support vector machine (SVM). The MARS model was applied on the daily output of a grid-connected 2.1kW PV system to provide the 1-day-ahead mean daily forecast of the power output. The comparisons with a wide variety of forecast models show that the MARS model is able to provide reliable forecast performance.

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

[2]  P. J. García Nieto,et al.  Nonlinear air quality modeling using multivariate adaptive regression splines in Gijón urban area (Northern Spain) at local scale , 2014, Appl. Math. Comput..

[3]  Yue Zhang,et al.  Day-Ahead Power Output Forecasting for Small-Scale Solar Photovoltaic Electricity Generators , 2015, IEEE Transactions on Smart Grid.

[4]  Yan Su,et al.  Real-time prediction models for output power and efficiency of grid-connected solar photovoltaic systems , 2012 .

[5]  R. Lewis.  An Introduction to Classification and Regression Tree (CART) Analysis , 2000 .

[6]  Johan A. K. Suykens,et al.  Least Squares Support Vector Machine Classifiers , 1999, Neural Processing Letters.

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

[8]  Seoung Bum Kim,et al.  A convex version of multivariate adaptive regression splines , 2015, Comput. Stat. Data Anal..

[9]  Wengang Zhang,et al.  Assessment of soil liquefaction based on capacity energy concept and multivariate adaptive regression splines , 2015 .

[10]  Maria Grazia De Giorgi,et al.  Error analysis of hybrid photovoltaic power forecasting models: A case study of mediterranean climate , 2015 .

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

[12]  T. Takashima,et al.  Use of support vector regression and numerically predicted cloudiness to forecast power output of a photovoltaic power plant in Kitakyushu, Japan , 2012 .

[13]  Cyril Voyant,et al.  Bayesian rules and stochastic models for high accuracy prediction of solar radiation , 2013, ArXiv.

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

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

[16]  Tomislav Šarić,et al.  Comparison of static and adaptive models for short-term residential natural gas forecasting in Croatia , 2014 .

[17]  Chao-Ming Huang,et al.  A Weather-Based Hybrid Method for 1-Day Ahead Hourly Forecasting of PV Power Output , 2014, IEEE Transactions on Sustainable Energy.

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

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

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

[21]  Upmanu Lall,et al.  A k‐nearest‐neighbor simulator for daily precipitation and other weather variables , 1999 .

[22]  J. Friedman.  Multivariate adaptive regression splines , 1990 .

[23]  Yongping Yang,et al.  Forecasting Power Output of Photovoltaic Systems Based on Weather Classification and Support Vector Machines , 2012 .

[24]  Yan Su,et al.  Analysis of daily solar power prediction with data-driven approaches , 2014 .

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

[26]  Vishwamitra Oree,et al.  A hybrid method for forecasting the energy output of photovoltaic systems , 2015 .

[27]  Inci Batmaz,et al.  A computational approach to nonparametric regression: bootstrapping CMARS method , 2015, Machine Learning.

[28]  Badia Amrouche,et al.  Artificial neural network based daily local forecasting for global solar radiation , 2014 .

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

[30]  Hasimah Abdul Rahman,et al.  A Novel Hybrid Model for Short-Term Forecasting in PV Power Generation , 2014 .

[31]  L. D. Monache,et al.  An analog ensemble for short-term probabilistic solar power forecast , 2015 .

[32]  Eduardo F. Fernández,et al.  A methodology based on dynamic artificial neural network for short-term forecasting of the power output of a PV generator , 2014 .

[33]  Ping-Feng Pai,et al.  Solar power output forecasting using evolutionary seasonal decomposition least-square support vector regression , 2016 .

[34]  Yan Su,et al.  An ARMAX model for forecasting the power output of a grid connected photovoltaic system , 2014 .

[35]  Jianzhou Wang,et al.  The study and application of a novel hybrid forecasting model – A case study of wind speed forecasting in China , 2015 .

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

[37]  O. Perpiñán,et al.  PV power forecast using a nonparametric PV model , 2015 .

[38]  K. Uma Rao,et al.  Development of statistical time series models for solar power prediction , 2015 .

[39]  Francesco Grimaccia,et al.  A Physical Hybrid Artificial Neural Network for Short Term Forecasting of PV Plant Power Output , 2015 .

[40]  Li-Yen Chang,et al.  Analysis of bilateral air passenger flows: A non-parametric multivariate adaptive regression spline approach , 2014 .