Hourly solar irradiance forecasting based on machine learning models

Size: px
Start display at page:

Download "Hourly solar irradiance forecasting based on machine learning models"

Transcription

1 th IEEE International Conference on Machine Learning and Applications Hourly solar irradiance forecasting based on machine learning models Fateh Nassim MELZI IRT SystemX, Palaiseau, France Taieb TOUATI, Allou SAME and Latifa OUKHELLOU Université Paris-Est, IFSTTAR, COSYS, GRETTIA, F Marne-la-Vallée, France (taieb.touati, allou.same, Abstract In recent years, many research studies are conducted into the use of smart meters data for developping decisionmaking tools including both analytical, forecasting and display purposes. Forecasting energy generation or forecasting energy consumption demand are indeed central problems for urban stakeholders (electricity companies and urban planners). These issues are helpful to allow them ensuring an efficient planning and optimization of energy resources. This paper investigates the problem for forecasting the hourly solar irradiance within a Machine Learning (ML) framework using Similarity method (SIM), Support Vector Machine (SVM) and Neural Network (NN). These approaches rely on a methodology which takes into account the previous hours of the predicting day and also the days having the same number of sunshine hours in the history. The study is conducted on a real data set collected on the Paris suburb of Alfortville. A comparison with two time series approaches namely Naive method and Autoregressive Moving Average Model (ARMA) is performed. This study is the first step towards the development of the hourly solar irradiance forecasting hybrid models. I. INTRODUCTION Nowadays, there is a growing interest on renewable energies. This trend is likely to continue with the need to address environmental issues and thanks to the decreased costs and technological advances of the renewable energy sources. Within this context, public stakeholders involving electricity suppliers, distribution network operators and city managers have to face a rapide increase in electricity demand, to diversify the sources of energy production and to be able to balance production, supply and demand. For these purposes, smart meters are increasingly used and the amount of data they collect are analysed. The aim is to build decision-making tools in order to be able to plan the electricity requirements depending on current and emerging needs. Therefore, photovoltaic energy is among the top three renewable energy used world wide. However, due to the fact that the output power depends on solar irradiance and weather conditions, one of the challenge to face to make this kind of energy production more attractive relies on the ability to develop advanced and precise irradiance solar measurements, modeling and forecasting. In so doing, a better planning and scheduling of different types of generation plants can be achieved. Considering the specific problem of solar irradiance forecasting, a variety of methods and tools can be handled depending on the relevant temporal and spatial prediction horizons as well as on the contextual variables available to make this prediction. Multivariate time series or machine learning methods can thus be developed to tackle this problem. This paper investigates the opportunity to explore three machine learning techniques namely: Similarity method, Support Vector Machine (SVM) and Neural Network (NN), in order to perform the hourly solar irradiance forecasting. These methods rely on a methodology which takes into account the previous hours of the predicting day and also the days in the history having the same number of sunshine hours. A comparison with two time series methods specifically: Naive method and Autoregressive Moving Average Model (ARMA) is performed. Experiments are carried out on a real data set collected over twelve years on the Paris suburb of Alfortville and provided by Reuniwatt company 1. The remainder of the paper is organized as follows. A review of related work is provided in Section II. The solar irradiance data set used throughout our study is described in Section III. The hourly forecasting methods are presented in Section IV. Experimental results are detailed and discussed in Section V. Section VI concludes the paper and gives some perspectives of further works. II. STATE OF THE ART Global horizontal irradiance (GHI) forecasting is the key step in the majority of photovoltaic prediction systems, thus an increase of interest among researchers is observed. The GHI forecast gives a more precise modeling and forecasting for solar power. Various solar irradiance forecast methodologies have been developed for different time horizons up to 24h [1]. We can divide the models into two main groups: statistical models and numerical weather predictions models. Statistical models stand on the analysis of historical data and are divided into two categories namely Time series models and Machine learning models. Neural networks (NNs) are considered as a reference method for solving complex problems in various fields and this is the reason why applications of NNs in the field of solar forecasting is mentioned in several scientific works [2], [3], [4]. In fact, the scope of historical data on meteorological databases and the fact that NNs are intrinsically data driven approaches make them very attractive for the /16 $ European Union DOI /ICMLA

2 scientific community. Most publications are moving towards the hour ahead solar forecast, as it is the prevalent operational forecast solicited by most of the utility companies in the insular grids field. However, Mellit et al. [5] proposed a multiplayer model for a daily irradiance forecasting, using as inputs both the mean daily irradiance and the mean daily temperature. Another popular machine learning method is the Support Vector Machines (SVMs) [6]. The use of these techniques in the field of solar energy is relatively recent [7], [8], [9]. Li et al. [10] compared SVMs regression and Hidden Markov Models [11], in order to forecast short-term solar irradiance. The algorithms have been implemented under three different weather conditions with the same datasets. Li et al. developed a Matlab interface, the weather forecasting platform to enhance solar forecasting. The spectrum of methods used in irradiance forecasting, can range from machine learning methods to linear statistical methods like the Autoregressive model (AR) and the Autoregressive moving average model (ARMA) [12]. These methods are efficient for predicting the future value of auto-correlated time series data. The popularity of the ARMA method within the scientific community lies in its capacity of extracting interesting statistical tools, in addition to the adoption of the well known Box-Jenkins method [13]. To set up an ARMA model, a necessary condition is the stationarity of the time series [14]. Ji et al. [15] use the Augmented Dickey-Fuller (ADF) test to measure the stationarity of the time series. As the original solar irradiance data is not stationary (daily and annual seasonalities), a pre-processing is essential. Lauret et al. [16] proposed a bechmarking of several machine learning techniques and an AR model. In order to make the solar irradiance times series stationary, they used the Bird clear sky model [17] to normalize the data. More recently, David et al. [18] used the same pre-processing to forecast solar irradiance with recursive ARMA-GARCH models. Besides its benefits on stationarity, this data normalization facilitates the learning process for Machine learning methods. In this paper, we focus on forecasting the hourly solar irradiance using three machine learning techniques which take into account the previous hours of the predicting day, and also the days in the history having the same number of hours of sunshine. III. DATA SET PRESENTATION A. Description of the Data Set The available data set consists of an hourly solar irradiance of the Paris suburb of Alfortville provided by Reuniwatt company 2. The Irradiance expressed in Watts/m2 is collected for a duration of 144 months (1st January 2004 to 31st December 2015). The data set is represented by a time series X having a dimension (D T ), where D is the number of days (4383 days) and T =24is the number of measurements recorded every day. It is worth noting that no missing values are detected in X. Figure 1 shows the daily solar irradiance curves for the year At first glance, we can distinguish three parts in each day: the first one represents the night before 2 the sunrise, the second is the part of the day between the sunrise and the sunset, while the third one corresponds to the night after the sunset. As it can be expected, the durations of these parts vary from one period to another in the year depending on the seasonal fluctuations. We explore the data through the elementary statistics presented in the next section. Fig. 1: Daily solar irradiance curves during the year 2015 B. Elementary Statistics Elementary statistics are first applied to have an overview on the data. Figure 2 represents the evolution of the number of hours of sunshine over the 12 years (from 2004 to 2015). We remark that the progression of the number of hours of sunshine is the same from one non-leap year to another and even for the leap years. These observations will help us in the forecast process detailed in the next section. When we focus on each type of year, Figure 3 shows the distribution of the number of hours of sunshine during the years 2012 and 2015 corresponding to a leap and non-leap years respectively. We find that the year is divided into 9 groups of days according to their number of hours of sunshine during the day. This number varies between 9 and 17 hours. Figure 4 represents the average irradiance profiles for each group during the year As it can be expected, the more the number of hours of sunshine is high, the greater the level of irradiance is. This is explained by the presence of the sunny days during the hot periods when the days are long, and contrary for the cloudy days during the cold periods. Using these information, in the next section, we will focus on forecasting hourly solar irradiance. Fig. 2: Evolution of the number of hours of sunshine over 12 years 442

3 (a) (b) Fig. 3: (a) Distribution of the number of hours of sunshine during a leap year (2012); (b) non-leap year 2015 Irradiance (W/m2) Number of hours Time (hour) Fig. 4: Average solar irradiance profiles for each group IV. FORECASTING METHODS In this section, we are interested in the two main categories of forecasting methods to predict the hourly solar irradiance. On the one hand, there are the times series models. On the other hand, we find machine learning techniques. They are succinctly described in the following. A. Time series models In this part we consider two time series approaches namely: Naive method and Autoregressive Moving Average model (ARMA), in order to forecast one hour ahead solar irradiance. These methods are described in the following. a) Naive method: is a simple forecasting method which is often used as benchmark to evaluate the performance of other prediction methods. It assumes that the predicted value is equal to the last observed one. In our case, the prediction of the solar irradiance at the hour h is expressed as follows: îr h = ir h 1, (1) where h =2,...,H, such as H =24. b) Autoregressive Moving Average (ARMA): is a linear stochastic process which describes a weakly stationary process through two polynomials. This model is a useful tool for forecasting time series. If X t N is a time series, ɛ t N a white noise, and L the lag operator then an ARMA(p,q) model can be written: Φ(L)X t =Θ(L)ɛ t, (2) where Φ and Θ are polynomials of degrees p and q respectively. Times series forecasting models needs stationarity [19] to be effective. Using the ADF-Test, we found that the solar irradiance time series is not stationary. In order to stationarize them, we use an analogue model to the one used by Lauret et al. [16]. More precisely, we compute the clear sky index k, which is the quotient of the measured irradiance X and the irradiance under a clear sky X clear as the following: k = X/X clear. (3) X clear is the output of the McClear model [20], available on the SoDa website 3. To forecast the clear sky index, we propose a unified number of predicting hours H by days in each month to guarantee a measured solar irradiance for the both time series. This filtering method removes about 1% of the annual total solar energy. H is variable for each month and bounded by a morning and an evening hours as illustrated in TableII. Month Morning hour Evening hour January 9 15 February 9 16 March 8 17 April 8 17 May 7 19 June 6 20 July 6 20 August 7 19 September 7 18 October 7 17 November 8 16 December 9 15 TABLE I: Morning and evening hours for each month The forecast is obtained by modeling the clear sky index time series through an ARMA(3,1) model

4 k h = α α i k h i + β ɛ h 1 (4) The solar irradiance at hour h is given by: where h =2,...,H. îr h = k h X h clear, (5) B. Machine Learning Models In this part, we target to forecast the hourly solar irradiance, using three machine learning techniques namely: Similarity method (SIM), Support Vector Machine (SVM) and Neural Network (NN). These methods are succinctly described in the following. The use of these methods relies on a methodology which takes into account the previous hours of the forecasting day, and also the days in the history with the same number of hours of sunshine as the one to be predicted. As described in section 3 (b), the number of hours of sunshine during the days is known both for leap and non-leap years. To forecast the hourly solar irradiance, we focus only on predicting the last H 1 sunshine hours, where H {9, 10,..., 17}. Each hour h =2,...,H is forecasted by using a model with h 1 inputs (previous hours). The methodology is described in Figure 5. E = {x h i,yh i }n is the training data set used to train the models, where n is the number of days with the same number of hours of sunshine as the day to predict. x h i is a vector with h 1 past values. The length of x h i is variable according to the forecasted hour h. yi h corresponds to the solar irradiance value at h. Similarly, (x h,y h ) is the test observation. For our study, we present now the methods that we have retained. ( wi h =exp 1 ( x h x h i )) 2 Cσ(x h,x h i ), (6) where σ is the standard deviation and C is constant fixed at The predicted value of the irradiance at the hour h is calculated as follows: îr h = n wi h yi h. (7) d) Support Vector Machines (SVMs): were invented by Cortes and Vapnik [21], in order to solve the discriminant and the regression problems. This machine learning technique aims at maximizing the margin between the hyperplane and the data, and at the same time, minimizing the empirical risk (as a cost function). In regression and time series forecasting applications, good performances were obtained [22]. For our application, the prediction of an input test case x h is given by: îr h = n α i f(x h,x h i )+b, (8) where f denotes the radial basis function, and b is a bias parameter. e) Neural Networks (NNs): A NN is a mathematical model that is inspired of the biological neural networks operation [23]. To define a non linear parameterized mapping from an input x i to an output y i, we design a NN that uses h 1 inputs, we fix the number of hidden neurons to 10 and we consider one single output (hour to forecast). The forecast of the input test variable x h is given by the following equation: îr h = m j=1 h 1 w j f( w ji x h + b 1 )+b 2, (9) the m hidden neurons are associated to the tangent hyperbolic function f. The weights w j, w ji and the biases b 1, b 2 are estimated during the training phase. It can be noticed that the number of hidden neurons can be optimized to improve the test performances of the model, which would be the subject of future investigation. Fig. 5: Methodology applied to forecast hourly solar irradiance using machine learning techniques c) Similarity method (SIM): is a simple method which is based on the comparison between the actual observation and the past observations in the history in order to forecast the future one. This method takes into account the hypothesis that the future following two similar observations are expected to be comparable. To evaluate how close is the test input variable x h with the past observations x h i in the history, a weight wi h is attributed to each observation x h i as follows: V. RESULTS AND DISCUSSION Two years of data (2014 and 2015) are considered to evaluate the performances of the models. The leave-one-out cross validation method is used, meaning that the training is performed on the n 1 past observations and the test is performed on the n th observation. This operation with different training and test sets is repeated n times, from 1st January 2015 to 31st December A statistical error indicator, namely the Normalized Root Mean Square Error (NRMSE) is computed to evaluate the accuracy of each model. This indicator is computed by determining the Root Mean Square Error (RMSE): 444

5 RMSE = 1 H (îr h ir h ) H 1 2, (10) h=2 where îr h and ir h are the hourly forecasted and measured irradiance respectively. H 1 is the number of hours to be predicted. Then we compute the NRMSE which is expressed by: NRMSE = RMSE, (11) īr where īr represents the mean measured irradiance corresponding to H 1 hours. Irradiance (W/m2) Irradiance (W/m2) Measured Naive ARMA SIM SVM NN Time (hour) (a) Measured Naive ARMA SIM SVM NN Time (hour) (b) Fig. 6: (a) Measured and forecasted values for a sunny day; (b) cloudy day Error indicator Models Naive ARMA SIM SVM NN NRMSE TABLE II: Normalized Root Mean Square Error for the hourly forecasting models After applying the five forecasting models, Figure 6 represents the measured and forecasted values using these methods for a sunny and a cloudy days respectively. For a sunny day (see Figure 6 (a)), we remark that the predicted values with ARMA, SIM, SVM and NN have well agreement with the measured ones. Especially, the forecasted solar irradiances of the three machine learning techniques are very close to the observed values comparing to the ARMA model. About the Naive method, the forecasted values are one hour lagged comparing to the measured ones. However, for the cloudy day (see Figure 6 (b)), the predicted values of the whole used models, follow the variations and the fluctuations of the measured irradiances, with a different adjustment. This difference is significant for Naive and SIM compared to the remaining methods. Our proposed models are accurate to forecast the sunny days without intermittency and less efficient in predicting the days with changeable weather. In the case of forecasting the cloudy days, it would be useful to integrate the meteorological variables (humidity, wind, temperature,...) in the models to increase their performances. Comparing the statistical error indicators of the five models (see Table II), we remark that ARMA, SIM, SVM and NN give better performances than the benchmark method (Naive). We observe also, that the NRMSE of ARMA is equal to that obtained by SVM and it is better than the SIM one. In this study, the best forecasting method is the NN with a NRMSE of The machine learning techniques can be more accurate by augmenting the training period. These performances are due to the methodology applied, which takes into account the previous hours of the forecasting day, and also the days having the same number of sunshine hours in the history. To perform a global evaluation of the models, we represent in Figure 7 the evolution of the NRMSE of the four forecasting methods according to the days of the year For certain extremely bad weather condition, machine learning techniques present a limitation in the forecasting comparing to ARMA. For example, during February 20th 2015, the machine learning techniques SIM, SVM and NN give a high NRMSE values: 1.627, and respectively, however ARMA obtains an acceptable NRMSE of To remedy this problem and take benefit from either machine learning and times series models, hybrid models could be a solution to extract full characteristics of solar irradiance. VI. CONCLUSION In this paper, an hourly solar irradiance forecasting methods have been used through three machine learning techniques, namely Similarity method (SIM), Support Vector Machine (SVM) and Neural Networks (NN). These methods rely on a methodology, which takes into account the previous hours of the forecasting day, and also the days in the history having the same number of sunshine hours. The study is conducted on a real data set collected on the Paris suburb of Alfortville. A comparison with two time series methods is performed, namely Naive method and Autoregressive Moving Average Model (ARMA). The proposed models give a satisfactory results during a sunny days and lesser accuracy for a cloudy days, thus it would be useful to integrate a meteorological variables (humidity, wind, temperature,...) to improve models performances. To take benefit from either machine learning and times series approaches, hybrids models could be a solution to extract full characteristics of solar irradiance. As future works, combination between hybrid models and meteorological variables are to be studied and investigated to improve the reliability for the extremely bad weather. 445

6 (a) (b) (c) (d) Fig. 7: (a) Evolution of the NRMSE according to the days of the year 2015 for Autoregressive Moving Average (ARMA); (b) Similarity method (SIM); (c) Support Vector Machine (SVM); (d) Neural Network (NN) [3] R. H. Inman, H. T. Pedro, and C. F. Coimbra, Solar forecasting methods for renewable energy integration, Progress in energy and combustion science, vol. 39, no. 6, pp , [4] A. Mellit, S. Kalogirou, L. Hontoria, and S. Shaari, Artificial intelligence techniques for sizing photovoltaic systems: A review, Renewable and Sustainable Energy Reviews, vol. 13, no. 2, pp , [5] A. Mellit and A. M. Pavan, A 24-h forecast of solar irradiance using artificial neural network: Application for performance prediction of a grid-connected pv plant at trieste, italy, Solar Energy, vol. 84, no. 5, pp , [6] V. Vapnik, The nature of statistical learning theory. Springer Science & Business Media, [7] J. Zeng and W. Qiao, Short-term solar power prediction using a support vector machine, Renewable Energy, vol. 52, pp , [8] B. Wolff, E. Lorenz, and O. Kramer, Statistical learning for shortterm photovoltaic power predictions, in Computational Sustainability. Springer, 2016, pp [9] J. G. d. S. Fonseca Jr, O. Takashi, O. Hideaki, S. Ken-ichi, T. Takumi, and O. Kazuhiko, Analysis of different techniques to set support vector regression to forecast insolation in tsukuba, japan, Journal of International Council on Electrical Engineering, vol. 3, no. 2, pp , [10] J. Li, J. K. Ward, J. Tong, L. Collins, and G. Platt, Machine learning for solar irradiance forecasting of photovoltaic system, Renewable Energy, vol. 90, pp , [11] R. J. Elliott, L. Aggoun, and J. B. Moore, Hidden Markov models: estimation and control. Springer Science & Business Media, 2008, vol. 29. [12] R. J. Hyndman and G. Athanasopoulos, Forecasting: principles and practice. OTexts, [13] J. Boland, Time series modelling of solar radiation, in Modeling Solar Radiation at the Earths Surface. Springer, 2008, pp [14] J. D. Hamilton, Time series analysis. Princeton university press Princeton, 1994, vol. 2. [15] W. Ji and K. C. Chee, Prediction of hourly solar radiation using a novel hybrid model of arma and tdnn, Solar Energy, vol. 85, no. 5, pp , [16] P. Lauret, C. Voyant, T. Soubdhan, M. David, and P. Poggi, A benchmarking of machine learning techniques for solar radiation forecasting in an insular context, Solar Energy, vol. 112, pp , [17] R. E. Bird and R. L. Hulstrom, Simplified clear sky model for direct and diffuse insolation on horizontal surfaces, Solar Energy Research Inst., Golden, CO (USA), Tech. Rep., [18] M. David, F. Ramahatana, P.-J. Trombe, and P. Lauret, Probabilistic forecasting of the solar irradiance with recursive arma and garch models, Solar Energy, vol. 133, pp , [19] S. Makridakis, S. C. Wheelwright, and R. J. Hyndman, Forecasting methods and applications. John Wiley & Sons, [20] M. Lefevre, A. Oumbe, P. Blanc, B. Espinar, B. Gschwind, Z. Qu, L. Wald, M. S. Homscheidt, C. Hoyer-Klick, A. Arola et al., Mcclear: a new model estimating downwelling solar radiation at ground level in clear-sky conditions, Atmospheric Measurement Techniques, vol. 6, pp , [21] C. Cortes and V. Vapnik, Support-vector networks, in Machine Learning, 1995, pp [22] K.-R. Müller, A. Smola, G. Rätsch, B. Schölkopf, J. Kohlmorgen, and V. Vapnik, Predicting time series with support vector machines, in Artificial Neural Networks: ICANN 97, Max-Planck-Gesellschaft. Berlin, Germany: Springer, Oct. 1997, pp [23] D. E. Rumelhart, G. E. Hinton, and R. J. Williams, Learning representations by back-propagating errors, Cognitive modeling, vol. 5, no. 3, p. 1, REFERENCES [1] M. Diagne, M. David, P. Lauret, J. Boland, and N. Schmutz, Review of solar irradiance forecasting methods and a proposition for small-scale insular grids, Renewable and Sustainable Energy Reviews, vol. 27, pp , [2] S. Quaiyum, S. Rahman, and S. Rahman, Application of artificial neural network in forecasting solar irradiance and sizing of photovoltaic cell for standalone systems in bangladesh, International Journal of Computer Applications, vol. 32, no. 10,

COMPARISON OF CLEAR-SKY MODELS FOR EVALUATING SOLAR FORECASTING SKILL

COMPARISON OF CLEAR-SKY MODELS FOR EVALUATING SOLAR FORECASTING SKILL COMPARISON OF CLEAR-SKY MODELS FOR EVALUATING SOLAR FORECASTING SKILL Ricardo Marquez Mechanical Engineering and Applied Mechanics School of Engineering University of California Merced Carlos F. M. Coimbra

More information

Short Term Solar Radiation Forecast from Meteorological Data using Artificial Neural Network for Yola, Nigeria

Short Term Solar Radiation Forecast from Meteorological Data using Artificial Neural Network for Yola, Nigeria American Journal of Engineering Research (AJER) 017 American Journal of Engineering Research (AJER) eiss: 300847 piss : 300936 Volume6, Issue8, pp8389 www.ajer.org Research Paper Open Access Short Term

More information

The Copernicus Atmosphere Monitoring Service (CAMS) Radiation Service in a nutshell

The Copernicus Atmosphere Monitoring Service (CAMS) Radiation Service in a nutshell The Copernicus Atmosphere Monitoring Service (CAMS) Radiation Service in a nutshell Issued by: M. Schroedter-Homscheidt, DLR Date: 17/02/2016 REF.: Service Contract No 2015/CAMS_72/SC1 This document has

More information

SOLAR ENERGY FORECASTING A PATHWAY FOR SUCCESSFUL RENEWABLE ENERGY INTEGRATION. (An ANN Based Model using NARX Model for forecasting of GHI)

SOLAR ENERGY FORECASTING A PATHWAY FOR SUCCESSFUL RENEWABLE ENERGY INTEGRATION. (An ANN Based Model using NARX Model for forecasting of GHI) IPS 2018 SOLAR ENERGY FORECASTING A PATHWAY FOR SUCCESSFUL RENEWABLE ENERGY INTEGRATION ABSTRACT (An ANN Based Model using NARX Model for forecasting of GHI) Manish Kumar Tikariha, DGM(Operations), NTPC

More information

A Novel Method for Predicting the Power Output of Distributed Renewable Energy Resources

A Novel Method for Predicting the Power Output of Distributed Renewable Energy Resources A Novel Method for Predicting the Power Output of Distributed Renewable Energy Resources Athanasios Aris Panagopoulos1 Supervisor: Georgios Chalkiadakis1 Technical University of Crete, Greece A thesis

More information

Soleksat. a flexible solar irradiance forecasting tool using satellite images and geographic web-services

Soleksat. a flexible solar irradiance forecasting tool using satellite images and geographic web-services Soleksat a flexible solar irradiance forecasting tool using satellite images and geographic web-services Sylvain Cros, Mathieu Turpin, Caroline Lallemand, Quentin Verspieren, Nicolas Schmutz They support

More information

Short-term wind forecasting using artificial neural networks (ANNs)

Short-term wind forecasting using artificial neural networks (ANNs) Energy and Sustainability II 197 Short-term wind forecasting using artificial neural networks (ANNs) M. G. De Giorgi, A. Ficarella & M. G. Russo Department of Engineering Innovation, Centro Ricerche Energia

More information

Day-Ahead Solar Forecasting Based on Multi-level Solar Measurements

Day-Ahead Solar Forecasting Based on Multi-level Solar Measurements Day-Ahead Solar Forecasting Based on Multi-level Solar Measurements Mohana Alanazi, Mohsen Mahoor, Amin Khodaei Department of Electrical and Computer Engineering University of Denver Denver, USA mohana.alanazi@du.edu,

More information

Short- term solar forecas/ng with sta/s/cal models and combina/on of models

Short- term solar forecas/ng with sta/s/cal models and combina/on of models Short- term solar forecas/ng with sta/s/cal models and combina/on of models Philippe Lauret, Mathieu David PIMENT Laboratory, University of La Réunion L. Mazorra Aguiar University of Las Palmas de Gran

More information

SEASONAL MODELING OF HOURLY SOLAR IRRADIATION SERIES

SEASONAL MODELING OF HOURLY SOLAR IRRADIATION SERIES SEASONAL MODELING OF HOURLY SOLAR IRRADIATION SERIES M. PAULESCU 1, N. POP 2, N. STEFU 1, E. PAULESCU 1, R. BOATA 3, D. CALINOIU 2 1 West University of Timisoara, Physics Department, V. Parvan 4, 300223

More information

Satellite-based solar irradiance assessment and forecasting in tropical insular areas

Satellite-based solar irradiance assessment and forecasting in tropical insular areas Satellite-based solar irradiance assessment and forecasting in tropical insular areas Sylvain Cros, Maxime De Roubaix, Mathieu Turpin, Patrick Jeanty 16th EMS Annual Meeting & 11th European Conference

More information

ANN and Statistical Theory Based Forecasting and Analysis of Power System Variables

ANN and Statistical Theory Based Forecasting and Analysis of Power System Variables ANN and Statistical Theory Based Forecasting and Analysis of Power System Variables Sruthi V. Nair 1, Poonam Kothari 2, Kushal Lodha 3 1,2,3 Lecturer, G. H. Raisoni Institute of Engineering & Technology,

More information

Solar Nowcasting with Cluster-based Detrending

Solar Nowcasting with Cluster-based Detrending Solar Nowcasting with Cluster-based Detrending Antonio Sanfilippo, Luis Pomares, Daniel Perez-Astudillo, Nassma Mohandes, Dunia Bachour ICEM 2017 Oral Presentation 26-29June 2017, Bari, Italy Overview

More information

Country scale solar irradiance forecasting for PV power trading

Country scale solar irradiance forecasting for PV power trading Country scale solar irradiance forecasting for PV power trading The benefits of the nighttime satellite-based forecast Sylvain Cros, Laurent Huet, Etienne Buessler, Mathieu Turpin European power exchange

More information

An application of MDLPF models for solar radiation forecasting

An application of MDLPF models for solar radiation forecasting International Journal of Smart Grid and Clean Energy An application of MDLPF models for solar radiation forecasting Emre Akarslan, Fatih Onur Hocaoglu * Afyon Kocatepe University, ANS Campus Engineering

More information

Deep Learning Architecture for Univariate Time Series Forecasting

Deep Learning Architecture for Univariate Time Series Forecasting CS229,Technical Report, 2014 Deep Learning Architecture for Univariate Time Series Forecasting Dmitry Vengertsev 1 Abstract This paper studies the problem of applying machine learning with deep architecture

More information

The document was not produced by the CAISO and therefore does not necessarily reflect its views or opinion.

The document was not produced by the CAISO and therefore does not necessarily reflect its views or opinion. Version No. 1.0 Version Date 2/25/2008 Externally-authored document cover sheet Effective Date: 4/03/2008 The purpose of this cover sheet is to provide attribution and background information for documents

More information

Probabilistic forecasting of the solar irradiance with recursive ARMA and GARCH models

Probabilistic forecasting of the solar irradiance with recursive ARMA and GARCH models Probabilistic forecasting of the solar irradiance with recursive ARMA and GARCH models M. David, F. Ramahatana, P.J. Trombe, Philippe Lauret To cite this version: M. David, F. Ramahatana, P.J. Trombe,

More information

Two-Stage Hybrid Day-Ahead Solar Forecasting

Two-Stage Hybrid Day-Ahead Solar Forecasting Two-Stage Hybrid Day-Ahead Solar Forecasting Mohana Alanazi, Mohsen Mahoor, Amin Khodaei Department of Electrical and Computer Engineering University of Denver Denver, USA mohana.alanazi@du.edu, mohsen.mahoor@du.edu,

More information

Evaluation of Time-Series, Regression and Neural Network Models for Solar Forecasting: Part I: One-Hour Horizon. Abstract

Evaluation of Time-Series, Regression and Neural Network Models for Solar Forecasting: Part I: One-Hour Horizon. Abstract Evaluation of Time-Series, Regression and Neural Network Models for Solar Forecasting: Part I: One-Hour Horizon Alireza Inanlougani, T.Agami Reddy and Srinivas Katiamula School of Computing, Informatics

More information

Post-processing of solar irradiance forecasts from WRF Model at Reunion Island

Post-processing of solar irradiance forecasts from WRF Model at Reunion Island Available online at www.sciencedirect.com ScienceDirect Energy Procedia 57 (2014 ) 1364 1373 2013 ISES Solar World Congress Post-processing of solar irradiance forecasts from WRF Model at Reunion Island

More information

A direct normal irradiation forecasting model based on artificial neural networks

A direct normal irradiation forecasting model based on artificial neural networks Revue des Energies Renouvelables Vol. 19 N 1 (016) 1 8 A direct normal irradiation forecasting model based on artificial neural networks I. Belhaj, O. El Fatni, S. Barhmi and E.H. Saidi Laboratory of High

More information

Solar irradiance forecasting for Chulalongkorn University location using time series models

Solar irradiance forecasting for Chulalongkorn University location using time series models Senior Project Proposal 2102499 Year 2016 Solar irradiance forecasting for Chulalongkorn University location using time series models Vichaya Layanun ID 5630550721 Advisor: Assist. Prof. Jitkomut Songsiri

More information

Estimation of Hourly Global Solar Radiation for Composite Climate

Estimation of Hourly Global Solar Radiation for Composite Climate Open Environmental Sciences, 28, 2, 34-38 34 Estimation of Hourly Global Solar Radiation for Composite Climate M. Jamil Ahmad and G.N. Tiwari * Open Access Center for Energy Studies, ndian nstitute of

More information

Short-Term Power Production Forecasting in Smart Grid Based on Solar Power Plants

Short-Term Power Production Forecasting in Smart Grid Based on Solar Power Plants International Journal of Engineering and Applied Sciences (IJEAS) Short-Term Power Production Forecasting in Smart Grid Based on Solar Power Plants Qudsia Memon, Nurettin Çetinkaya Abstract Since the world

More information

Analysis and validation of 24 hours ahead neural network forecasting of photovoltaic output power

Analysis and validation of 24 hours ahead neural network forecasting of photovoltaic output power 1 2 3 4 5 6 Analysis and validation of 24 hours ahead neural network forecasting of photovoltaic output power A. Dolara, F. Grimaccia, S. Leva, M. Mussetta, E. Ogliari Dipartimento di Energia, Politecnico

More information

Solar irradiance forecasting for Chulalongkorn University location using time series models

Solar irradiance forecasting for Chulalongkorn University location using time series models Senior Project Proposal 2102490 Year 2016 Solar irradiance forecasting for Chulalongkorn University location using time series models Vichaya Layanun ID 5630550721 Advisor: Assist. Prof. Jitkomut Songsiri

More information

Recurrent Neural Network based Deep Learning for Solar Radiation Prediction

Recurrent Neural Network based Deep Learning for Solar Radiation Prediction Recurrent Neural Network based Deep Learning for Solar Radiation Prediction Fuxin Niu, Zheng O Neill Department of Mechanical Engineering, The University of Alabama, Tuscaloosa, USA Abstract In the past

More information

Multi-Model Ensemble for day ahead PV power forecasting improvement

Multi-Model Ensemble for day ahead PV power forecasting improvement Multi-Model Ensemble for day ahead PV power forecasting improvement Cristina Cornaro a,b, Marco Pierro a,e, Francesco Bucci a, Matteo De Felice d, Enrico Maggioni c, David Moser e,alessandro Perotto c,

More information

Research Article Hybrid Power Forecasting Model for Photovoltaic Plants Based on Neural Network with Air Quality Index

Research Article Hybrid Power Forecasting Model for Photovoltaic Plants Based on Neural Network with Air Quality Index Hindawi International Photoenergy Volume 2017, Article ID 6938713, 9 pages https://doi.org/10.1155/2017/6938713 Research Article Hybrid Power Forecasting Model for Photovoltaic Plants Based on Neural Network

More information

About Nnergix +2, More than 2,5 GW forecasted. Forecasting in 5 countries. 4 predictive technologies. More than power facilities

About Nnergix +2, More than 2,5 GW forecasted. Forecasting in 5 countries. 4 predictive technologies. More than power facilities About Nnergix +2,5 5 4 +20.000 More than 2,5 GW forecasted Forecasting in 5 countries 4 predictive technologies More than 20.000 power facilities Nnergix s Timeline 2012 First Solar Photovoltaic energy

More information

COMPARISON OF CLEAR-SKY MODELS FOR EVALUATING SOLAR FORECASTING SKILL

COMPARISON OF CLEAR-SKY MODELS FOR EVALUATING SOLAR FORECASTING SKILL COMPARISON OF CLEAR-SKY MODELS FOR EVALUATING SOLAR FORECASTING SKILL Ricardo Marquez Mechanical Engineering Applied Mechanics School of Engineering University of California Merced Merced, California 95343

More information

IMPROVED MODEL FOR FORECASTING GLOBAL SOLAR IRRADIANCE DURING SUNNY AND CLOUDY DAYS. Bogdan-Gabriel Burduhos, Mircea Neagoe *

IMPROVED MODEL FOR FORECASTING GLOBAL SOLAR IRRADIANCE DURING SUNNY AND CLOUDY DAYS. Bogdan-Gabriel Burduhos, Mircea Neagoe * DOI: 10.2478/awutp-2018-0002 ANNALS OF WEST UNIVERSITY OF TIMISOARA PHYSICS Vol. LX, 2018 IMPROVED MODEL FOR FORECASTING GLOBAL SOLAR IRRADIANCE DURING SUNNY AND CLOUDY DAYS Bogdan-Gabriel Burduhos, Mircea

More information

GLOBAL SOLAR RADIATION MODELING ON A HORIZONTAL SURFACE USING POLYNOMIAL FITTING

GLOBAL SOLAR RADIATION MODELING ON A HORIZONTAL SURFACE USING POLYNOMIAL FITTING GLOBAL SOLAR RADIATION MODELING ON A HORIZONTAL SURFACE USING POLYNOMIAL FITTING Michael Gyan * and Abraham Amankwah University of Ghana - Department of physics ABSTRACT: An attempt has been made to use

More information

Multi-Plant Photovoltaic Energy Forecasting Challenge: Second place solution

Multi-Plant Photovoltaic Energy Forecasting Challenge: Second place solution Multi-Plant Photovoltaic Energy Forecasting Challenge: Second place solution Clément Gautrais 1, Yann Dauxais 1, and Maël Guilleme 2 1 University of Rennes 1/Inria Rennes clement.gautrais@irisa.fr 2 Energiency/University

More information

A Novel 2-D Model Approach for the Prediction of Hourly Solar Radiation

A Novel 2-D Model Approach for the Prediction of Hourly Solar Radiation A Novel 2-D Model Approach for the Prediction of Hourly Solar Radiation F Onur Hocao glu, Ö Nezih Gerek, and Mehmet Kurban Anadolu University, Dept of Electrical and Electronics Eng, Eskisehir, Turkey

More information

Predicting the Electricity Demand Response via Data-driven Inverse Optimization

Predicting the Electricity Demand Response via Data-driven Inverse Optimization Predicting the Electricity Demand Response via Data-driven Inverse Optimization Workshop on Demand Response and Energy Storage Modeling Zagreb, Croatia Juan M. Morales 1 1 Department of Applied Mathematics,

More information

Daily Global Solar Radiation estimation for Gran Canaria Island using Artificial Neural Networks

Daily Global Solar Radiation estimation for Gran Canaria Island using Artificial Neural Networks International Conference on Renewable Energies and Power Quality (ICREPQ 16) Madrid (Spain), 4 th to 6 th May, 2016 exçxãtuäx XÇxÜzç tçw céãxü dâtä àç ]ÉâÜÇtÄ (RE&PQJ) ISSN 2172-038 X, No.14 May 2016 Daily

More information

Review of solar irradiance forecasting methods and a proposition for small-scale insular grids

Review of solar irradiance forecasting methods and a proposition for small-scale insular grids Review of solar irradiance forecasting methods and a proposition for small-scale insular grids Hadja Maïmouna Diagne, Mathieu David, Philippe Lauret, John Boland, Nicolas Schmutz To cite this version:

More information

iafor The International Academic Forum

iafor The International Academic Forum Nowcasting of Global Horizontal Irradiance for an Equatorial-Based Location Using Artificial Neural Network and Machine Learning Kyairul Azmi Baharin, Universiti Teknologi Malaysia, Malaysia Hasimah Abd

More information

A benchmarking of machine learning techniques for solar radiation forecasting in an insular context

A benchmarking of machine learning techniques for solar radiation forecasting in an insular context A benchmarking of machine learning techniques for solar radiation forecasting in an insular context Philippe Lauret, Cyril Voyant, Ted Soubdhan, Mathieu David, Philippe Poggi To cite this version: Philippe

More information

Short-term water demand forecast based on deep neural network ABSTRACT

Short-term water demand forecast based on deep neural network ABSTRACT Short-term water demand forecast based on deep neural network Guancheng Guo 1, Shuming Liu 2 1,2 School of Environment, Tsinghua University, 100084, Beijing, China 2 shumingliu@tsinghua.edu.cn ABSTRACT

More information

Solar Power Forecasting Using Support Vector Regression

Solar Power Forecasting Using Support Vector Regression M. Abuella and B. Chowdhury, Solar Power Forecasting Using Support Vector Regression, in Proceedings of the American Society for Engineering Management International Annual Conference, 2016. Solar Power

More information

CHAPTER 6 CONCLUSION AND FUTURE SCOPE

CHAPTER 6 CONCLUSION AND FUTURE SCOPE CHAPTER 6 CONCLUSION AND FUTURE SCOPE 146 CHAPTER 6 CONCLUSION AND FUTURE SCOPE 6.1 SUMMARY The first chapter of the thesis highlighted the need of accurate wind forecasting models in order to transform

More information

OUTPUTS AND ERROR INDICATORS FOR SOLAR FORECASTING MODELS

OUTPUTS AND ERROR INDICATORS FOR SOLAR FORECASTING MODELS OUTPUTS AND ERROR INDICATORS FOR SOLAR FORECASTING MODELS Mathieu DAVID Hadja Maïmouna DIAGNE Philippe LAURET PIMENT University of La Reunion 97715 Saint Denis Cedex 9 Reunion Island mathieu.david@univ-reunion.fr

More information

Developing a Guide for Non-experts to Determine the Most Appropriate Use of Solar Energy Resource Information

Developing a Guide for Non-experts to Determine the Most Appropriate Use of Solar Energy Resource Information Developing a Guide for Non-experts to Determine the Most Appropriate Use of Solar Energy Resource Information Carsten Hoyer-Klick 1*, Jennifer McIntosh 2, Magda Moner-Girona 3, David Renné 4, Richard Perez

More information

Development of Short Term Solar Forecasts

Development of Short Term Solar Forecasts power systems eehlaboratory Seraina Buchmeier Development of Short Term Solar Forecasts Semester Project EEH Power Systems Laboratory Swiss Federal Institute of Technology (ETH) Zurich Supervisors: Olivier

More information

Study on Impact of Solar Photovoltaic Generation by Atmospheric Variables

Study on Impact of Solar Photovoltaic Generation by Atmospheric Variables Installed capacity in MW K. Sreedhara Babu et al., International Journal of Research in Engineering, IT and Social Sciences, Study on Impact of Solar Photovoltaic Generation by Atmospheric Variables K.

More information

MODELLING ENERGY DEMAND FORECASTING USING NEURAL NETWORKS WITH UNIVARIATE TIME SERIES

MODELLING ENERGY DEMAND FORECASTING USING NEURAL NETWORKS WITH UNIVARIATE TIME SERIES MODELLING ENERGY DEMAND FORECASTING USING NEURAL NETWORKS WITH UNIVARIATE TIME SERIES S. Cankurt 1, M. Yasin 2 1&2 Ishik University Erbil, Iraq 1 s.cankurt@ishik.edu.iq, 2 m.yasin@ishik.edu.iq doi:10.23918/iec2018.26

More information

SOLAR POWER FORECASTING BASED ON NUMERICAL WEATHER PREDICTION, SATELLITE DATA, AND POWER MEASUREMENTS

SOLAR POWER FORECASTING BASED ON NUMERICAL WEATHER PREDICTION, SATELLITE DATA, AND POWER MEASUREMENTS BASED ON NUMERICAL WEATHER PREDICTION, SATELLITE DATA, AND POWER MEASUREMENTS Detlev Heinemann, Elke Lorenz Energy Meteorology Group, Institute of Physics, Oldenburg University Workshop on Forecasting,

More information

Short-Term Load Forecasting Using ARIMA Model For Karnataka State Electrical Load

Short-Term Load Forecasting Using ARIMA Model For Karnataka State Electrical Load International Journal of Engineering Research and Development e-issn: 2278-67X, p-issn: 2278-8X, www.ijerd.com Volume 13, Issue 7 (July 217), PP.75-79 Short-Term Load Forecasting Using ARIMA Model For

More information

Iterative ARIMA-Multiple Support Vector Regression models for long term time series prediction

Iterative ARIMA-Multiple Support Vector Regression models for long term time series prediction and Machine Learning Bruges (Belgium), 23-25 April 24, i6doccom publ, ISBN 978-2874995-7 Available from http://wwwi6doccom/fr/livre/?gcoi=2843244 Iterative ARIMA-Multiple Support Vector Regression models

More information

Clustering and Prediction of Solar Radiation Daily Patterns

Clustering and Prediction of Solar Radiation Daily Patterns Int'l Conf. Data Mining DMIN'16 3 Clustering and Prediction of Solar Radiation Daily Patterns G. Nunnari 1, and S. Nunnari 1 1 Dipartimento di Ingegneria Elettrica, Elettronica e Informatica, Universitá

More information

DNICast Direct Normal Irradiance Nowcasting methods for optimized operation of concentrating solar technologies

DNICast Direct Normal Irradiance Nowcasting methods for optimized operation of concentrating solar technologies DNICast Direct Normal Irradiance Nowcasting methods for optimized operation of concentrating solar technologies THEME [ENERGY.2013.2.9.2] [Methods for the estimation of the Direct Normal Irradiation (DNI)]

More information

Feature Selection Optimization Solar Insolation Prediction Using Artificial Neural Network: Perspective Bangladesh

Feature Selection Optimization Solar Insolation Prediction Using Artificial Neural Network: Perspective Bangladesh American Journal of Engineering Research (AJER) 2016 American Journal of Engineering Research (AJER) e-issn: 2320-0847 p-issn : 2320-0936 Volume-5, Issue-8, pp-261-265 www.ajer.org Research Paper Open

More information

Multi-Plant Photovoltaic Energy Forecasting Challenge with Regression Tree Ensembles and Hourly Average Forecasts

Multi-Plant Photovoltaic Energy Forecasting Challenge with Regression Tree Ensembles and Hourly Average Forecasts Multi-Plant Photovoltaic Energy Forecasting Challenge with Regression Tree Ensembles and Hourly Average Forecasts Kathrin Bujna 1 and Martin Wistuba 2 1 Paderborn University 2 IBM Research Ireland Abstract.

More information

Probabilistic forecasting of solar radiation

Probabilistic forecasting of solar radiation Probabilistic forecasting of solar radiation Dr Adrian Grantham School of Information Technology and Mathematical Sciences School of Engineering 7 September 2017 Acknowledgements Funding: Collaborators:

More information

A Hybrid ARIMA and Neural Network Model to Forecast Particulate. Matter Concentration in Changsha, China

A Hybrid ARIMA and Neural Network Model to Forecast Particulate. Matter Concentration in Changsha, China A Hybrid ARIMA and Neural Network Model to Forecast Particulate Matter Concentration in Changsha, China Guangxing He 1, Qihong Deng 2* 1 School of Energy Science and Engineering, Central South University,

More information

SOLAR RADIATION FORECAST USING NEURAL NETWORKS FOR THE PREDICTION OF GRID CONNECTED PV PLANTS ENERGY PRODUCTION (DSP PROJECT)

SOLAR RADIATION FORECAST USING NEURAL NETWORKS FOR THE PREDICTION OF GRID CONNECTED PV PLANTS ENERGY PRODUCTION (DSP PROJECT) SOLAR RADIATION FORECAST USING NEURAL NETWORKS FOR THE PREDICTION OF GRID CONNECTED PV PLANTS ENERGY PRODUCTION (DSP PROJECT) C. Cornaro* #, F. Bucci*, M. Pierro*, F. Del Frate, S. Peronaci, A. Taravat

More information

Wind Resource Modelling using the Markov Transition Probability Matrix Method

Wind Resource Modelling using the Markov Transition Probability Matrix Method Twelfth LACCEI Latin American and Caribbean Conference for Engineering and Technology (LACCEI 214) Excellence in Engineering To Enhance a Country s Productivity July 22-24, 214 Guayaquil, Ecuador. Wind

More information

Bayesian Based Neural Network Model for Solar Photovoltaic Power Forecasting

Bayesian Based Neural Network Model for Solar Photovoltaic Power Forecasting Bayesian Based Neural Network Model for Solar Photovoltaic Power Forecasting Angelo Ciaramella 1, Antonino Staiano 1, Guido Cervone 2, and Stefano Alessandrini 3 1 Dept. of Science and Technology, University

More information

CONTROL AND OPTIMIZATION IN SMART-GRIDS

CONTROL AND OPTIMIZATION IN SMART-GRIDS CONTROL AND OPTIMIZATION IN SMART-GRIDS Fredy Ruiz Ph.D. Pontificia Universidad Javeriana, Colombia Visiting Profesor - ruizf@javeriana.edu.co May, 2018 Course topics Session 1: Introduction to Power systems

More information

Hybrid methodology for hourly global. radiation forecasting in Mediterranean area

Hybrid methodology for hourly global. radiation forecasting in Mediterranean area Hybrid methodology for hourly global radiation forecasting in Mediterranean area Cyril Voyant 1,2*, Marc Muselli 1, Christophe Paoli 1, Marie-Laure Nivet 1 1 University of Corsica, CNRS UMR SPE 6134, 20250

More information

Time Series and Forecasting

Time Series and Forecasting Time Series and Forecasting Introduction to Forecasting n What is forecasting? n Primary Function is to Predict the Future using (time series related or other) data we have in hand n Why are we interested?

More information

Temporal global solar radiation forecasting using artificial neural network in Tunisian climate

Temporal global solar radiation forecasting using artificial neural network in Tunisian climate Temporal global solar radiation forecasting using artificial neural network in Tunisian climate M. LOGHMARI Ismail #1, M.Youssef TIMOUMI *2 # Mechanical engineering laboratory, National Engineering School

More information

Frequency Forecasting using Time Series ARIMA model

Frequency Forecasting using Time Series ARIMA model Frequency Forecasting using Time Series ARIMA model Manish Kumar Tikariha DGM(O) NSPCL Bhilai Abstract In view of stringent regulatory stance and recent tariff guidelines, Deviation Settlement mechanism

More information

Pilot applications for Egypt related end-users

Pilot applications for Egypt related end-users GEO-CRADLE Regional Workshop Thursday, 25 th May, 2017 Pilot applications for Egypt related end-users Hesham El-Askary Chapman University Panagiotis Kosmopoulos National Observatory of Athens Stelios Kazadzis

More information

Time Series Analysis -- An Introduction -- AMS 586

Time Series Analysis -- An Introduction -- AMS 586 Time Series Analysis -- An Introduction -- AMS 586 1 Objectives of time series analysis Data description Data interpretation Modeling Control Prediction & Forecasting 2 Time-Series Data Numerical data

More information

ESTIMATION OF SOLAR RADIATION USING LEAST SQUARE SUPPORT VECTOR MACHINE

ESTIMATION OF SOLAR RADIATION USING LEAST SQUARE SUPPORT VECTOR MACHINE ESTIMATION OF SOLAR RADIATION USING LEAST SQUARE SUPPORT VECTOR MACHINE Sthita pragyan Mohanty Prasanta Kumar Patra Department of Computer Science and Department of Computer Science and Engineering CET

More information

That s Hot: Predicting Daily Temperature for Different Locations

That s Hot: Predicting Daily Temperature for Different Locations That s Hot: Predicting Daily Temperature for Different Locations Alborz Bejnood, Max Chang, Edward Zhu Stanford University Computer Science 229: Machine Learning December 14, 2012 1 Abstract. The problem

More information

Localized Online Weather Predictions with Overnight Adaption

Localized Online Weather Predictions with Overnight Adaption Localized Online Weather Predictions with Overnight Adaption Michael Zauner 1, Michaela Killian 1, and Martin Kozek 1 Vienna University of Technology, Institute of Mechanics and Mechatronics, Getreidemarkt

More information

A Support Vector Regression Model for Forecasting Rainfall

A Support Vector Regression Model for Forecasting Rainfall A Support Vector Regression for Forecasting Nasimul Hasan 1, Nayan Chandra Nath 1, Risul Islam Rasel 2 Department of Computer Science and Engineering, International Islamic University Chittagong, Bangladesh

More information

OPTIMIZATION OF GLOBAL SOLAR RADIATION OF TILT ANGLE FOR SOLAR PANELS, LOCATION: OUARGLA, ALGERIA

OPTIMIZATION OF GLOBAL SOLAR RADIATION OF TILT ANGLE FOR SOLAR PANELS, LOCATION: OUARGLA, ALGERIA OPTIMIZATION OF GLOBAL SOLAR RADIATION OF TILT ANGLE FOR SOLAR PANELS, LOCATION: OUARGLA, ALGERIA Mohamed Lakhdar LOUAZENE Dris KORICHI Department of Electrical Engineering, University of Ouargla, Algeria.

More information

Direct Normal Radiation from Global Radiation for Indian Stations

Direct Normal Radiation from Global Radiation for Indian Stations RESEARCH ARTICLE OPEN ACCESS Direct Normal Radiation from Global Radiation for Indian Stations Jaideep Rohilla 1, Amit Kumar 2, Amit Tiwari 3 1(Department of Mechanical Engineering, Somany Institute of

More information

Fine-grained Photovoltaic Output Prediction using a Bayesian Ensemble

Fine-grained Photovoltaic Output Prediction using a Bayesian Ensemble Fine-grained Photovoltaic Output Prediction using a Bayesian Ensemble 1,2, Manish Marwah 3,, Martin Arlitt 3, and Naren Ramakrishnan 1,2 1 Department of Computer Science, Virginia Tech, Blacksburg, VA

More information

A State Space Model for Wind Forecast Correction

A State Space Model for Wind Forecast Correction A State Space Model for Wind Forecast Correction Valrie Monbe, Pierre Ailliot 2, and Anne Cuzol 1 1 Lab-STICC, Université Européenne de Bretagne, France (e-mail: valerie.monbet@univ-ubs.fr, anne.cuzol@univ-ubs.fr)

More information

Bayesian rules and stochastic models for high accuracy prediction of solar radiation

Bayesian rules and stochastic models for high accuracy prediction of solar radiation Bayesian rules and stochastic models for high accuracy prediction of solar radiation Cyril Voyant 1*, Christophe Darras 2, Marc Muselli 2, Christophe Paoli 2, Marie-Laure Nivet 2 and Philippe Poggi 2 1-CHD

More information

COMPARISON OF MEASURED AND ESTIMATED SOLAR RADIATION DATA: A CASE STUDY FOR ISTANBUL

COMPARISON OF MEASURED AND ESTIMATED SOLAR RADIATION DATA: A CASE STUDY FOR ISTANBUL COMPARISON OF MEASURED AND ESTIMATED SOLAR RADIATION DATA: A CASE STUDY FOR ISTANBUL Şaban Pusat 1, Erdal Bağcı 2 1 Yildiz Technical University, Mechanical Engineering Department, Istanbul 2 Uzman Enerji

More information

FORECAST OF ENSEMBLE POWER PRODUCTION BY GRID-CONNECTED PV SYSTEMS

FORECAST OF ENSEMBLE POWER PRODUCTION BY GRID-CONNECTED PV SYSTEMS FORECAST OF ENSEMBLE POWER PRODUCTION BY GRID-CONNECTED PV SYSTEMS Elke Lorenz*, Detlev Heinemann*, Hashini Wickramarathne*, Hans Georg Beyer +, Stefan Bofinger * University of Oldenburg, Institute of

More information

Pilot applications for Greece and Egypt related end-users

Pilot applications for Greece and Egypt related end-users GEO-CRADLE Project Meeting 2 16 th November, 2016 Pilot applications for Greece and Egypt related end-users Panagiotis Kosmopoulos & Hesham El-Askary National Observatory of Athens Chapman University Eratosthenes

More information

Prediction of Hourly Solar Radiation in Amman-Jordan by Using Artificial Neural Networks

Prediction of Hourly Solar Radiation in Amman-Jordan by Using Artificial Neural Networks Int. J. of Thermal & Environmental Engineering Volume 14, No. 2 (2017) 103-108 Prediction of Hourly Solar Radiation in Amman-Jordan by Using Artificial Neural Networks M. A. Hamdan a*, E. Abdelhafez b

More information

CARLOS F. M. COIMBRA (PI) HUGO T. C. PEDRO (CO-PI)

CARLOS F. M. COIMBRA (PI) HUGO T. C. PEDRO (CO-PI) HIGH-FIDELITY SOLAR POWER FORECASTING SYSTEMS FOR THE 392 MW IVANPAH SOLAR PLANT (CSP) AND THE 250 MW CALIFORNIA VALLEY SOLAR RANCH (PV) PROJECT CEC EPC-14-008 CARLOS F. M. COIMBRA (PI) HUGO T. C. PEDRO

More information

22/04/2014. Economic Research

22/04/2014. Economic Research 22/04/2014 Economic Research Forecasting Models for Exchange Rate Tuesday, April 22, 2014 The science of prognostics has been going through a rapid and fruitful development in the past decades, with various

More information

Statistical methods for the prediction of night-time cooling and minimum temperature

Statistical methods for the prediction of night-time cooling and minimum temperature Meteorol. Appl. 13, 169 178 (26) doi:1.117/s1348276276 Statistical methods for the prediction of night-time cooling and minimum temperature G. Emmanouil 1, G. Galanis 1,2 & G. Kallos 1 1 University of

More information

Day Ahead Hourly Load and Price Forecast in ISO New England Market using ANN

Day Ahead Hourly Load and Price Forecast in ISO New England Market using ANN 23 Annual IEEE India Conference (INDICON) Day Ahead Hourly Load and Price Forecast in ISO New England Market using ANN Kishan Bhushan Sahay Department of Electrical Engineering Delhi Technological University

More information

An Adaptive Multi-Modeling Approach to Solar Nowcasting

An Adaptive Multi-Modeling Approach to Solar Nowcasting An Adaptive Multi-Modeling Approach to Solar Nowcasting Antonio Sanfilippo, Luis Martin-Pomares, Nassma Mohandes, Daniel Perez-Astudillo, Dunia A. Bachour ICEM 2015 Overview Introduction Background, problem

More information

The Research of Urban Rail Transit Sectional Passenger Flow Prediction Method

The Research of Urban Rail Transit Sectional Passenger Flow Prediction Method Journal of Intelligent Learning Systems and Applications, 2013, 5, 227-231 Published Online November 2013 (http://www.scirp.org/journal/jilsa) http://dx.doi.org/10.4236/jilsa.2013.54026 227 The Research

More information

Solar Irradiance Prediction using Neural Model

Solar Irradiance Prediction using Neural Model Volume-8, Issue-3, June 2018 International Journal of Engineering and Management Research Page Number: 241-245 DOI: doi.org/10.31033/ijemr.8.3.32 Solar Irradiance Prediction using Neural Model Raj Kumar

More information

Radial basis function neural networks model to estimate global solar radiation in semi-arid area

Radial basis function neural networks model to estimate global solar radiation in semi-arid area Leonardo Electronic Journal of Practices and Technologies ISSN 1583-1078 Issue 27, July-December 2015 p. 177-184 Radial basis function neural networks model to estimate global solar radiation in semi-arid

More information

USE OF FUZZY LOGIC TO INVESTIGATE WEATHER PARAMETER IMPACT ON ELECTRICAL LOAD BASED ON SHORT TERM FORECASTING

USE OF FUZZY LOGIC TO INVESTIGATE WEATHER PARAMETER IMPACT ON ELECTRICAL LOAD BASED ON SHORT TERM FORECASTING Nigerian Journal of Technology (NIJOTECH) Vol. 35, No. 3, July 2016, pp. 562 567 Copyright Faculty of Engineering, University of Nigeria, Nsukka, Print ISSN: 0331-8443, Electronic ISSN: 2467-8821 www.nijotech.com

More information

Solar Irradiance and Load Demand Forecasting based on Single Exponential Smoothing Method

Solar Irradiance and Load Demand Forecasting based on Single Exponential Smoothing Method Solar Irradiance and Load Demand Forecasting based on Single Exponential Smoothing Method P. Y. Lim and C. V. Nayar Abstract Forecasting of the solar irradiance and load demand are essential for system

More information

Improving the accuracy of solar irradiance forecasts based on Numerical Weather Prediction

Improving the accuracy of solar irradiance forecasts based on Numerical Weather Prediction Improving the accuracy of solar irradiance forecasts based on Numerical Weather Prediction Bibek Joshi, Alistair Bruce Sproul, Jessie Kai Copper, Merlinde Kay Why solar power forecasting? Electricity grid

More information

A comparative study of ANN and angstrom Prescott model in the context of solar radiation analysis

A comparative study of ANN and angstrom Prescott model in the context of solar radiation analysis A comparative study of ANN and angstrom Prescott model in the context of solar radiation analysis JUHI JOSHI 1, VINIT KUMAR 2 1 M.Tech, SGVU, Jaipur, India 2 Assistant. Professor, SGVU, Jaipur, India ABSTRACT

More information

Estimation of Diffuse Solar Radiation for Yola, Adamawa State, North- Eastern, Nigeria

Estimation of Diffuse Solar Radiation for Yola, Adamawa State, North- Eastern, Nigeria International Research Journal of Engineering and Technology (IRJET) e-issn: - Volume: Issue: Nov- www.irjet.net p-issn: - Estimation of Diffuse Solar Radiation for Yola, Adamawa State, North- Eastern,

More information

A Typical Meteorological Year for Energy Simulations in Hamilton, New Zealand

A Typical Meteorological Year for Energy Simulations in Hamilton, New Zealand Anderson T N, Duke M & Carson J K 26, A Typical Meteorological Year for Energy Simulations in Hamilton, New Zealand IPENZ engineering trenz 27-3 A Typical Meteorological Year for Energy Simulations in

More information

Using artificial neural network for solar energy level predicting

Using artificial neural network for solar energy level predicting ISSN 1746-7659, England, UK Journal of Information and Computing Science Vol. 13, No. 3, 2018, pp.163-167 Using artificial neural network for solar energy level predicting A.G. Mustafaev Dagestan state

More information

Modeling Of Global Solar Radiation By Using Ambient Air Temperature At Coastal Cities In India

Modeling Of Global Solar Radiation By Using Ambient Air Temperature At Coastal Cities In India International Journal of Applied Engineering Research ISSN 0973-4562 Volume 10, Number 7 (2015) pp. 16843-16852 Research India Publications http://www.ripublication.com Modeling Of Global Solar Radiation

More information

PROJECTING THE SOLAR RADIATION IN NASARAWA-NIGERIA USING REITVELD EQUATION

PROJECTING THE SOLAR RADIATION IN NASARAWA-NIGERIA USING REITVELD EQUATION PROJECTING THE SOLAR RADIATION IN NASARAWA-NIGERIA USING REITVELD EQUATION * Benedicta C. Njoku 1, Musibau A. Adesina 2, Eshipemi W. Lukman 3 and Bankole Omojola 4 1,2,3 Mechanical Engineering Department,

More information

Time Series and Forecasting

Time Series and Forecasting Time Series and Forecasting Introduction to Forecasting n What is forecasting? n Primary Function is to Predict the Future using (time series related or other) data we have in hand n Why are we interested?

More information

Climate services in support of the energy transformation

Climate services in support of the energy transformation services in support of the energy transformation EGU 11 April 2018, Vienna, Austria Climate Alberto Troccoli, Sylvie Parey, and the C3S ECEM team O u t l i n e Background of the C3S European Climatic Energy

More information