Elsevier

Automatica

Volume 113, March 2020, 108688
Automatica

Brief Paper
Estimation of photovoltaic generation forecasting models using limited information☆

https://doi.org/10.1016/j.automatica.2019.108688Get rights and content

Abstract

This work deals with the problem of estimating a photovoltaic generation forecasting model in scenarios where measurements of meteorological variables (i.e., solar irradiance and temperature) at the plant site are not available. A novel algorithm for the estimation of the parameters of the well-known PVUSA model of a photovoltaic plant is proposed. Such a method is characterized by a low computational complexity, and efficiently exploits only power generation measurements, a theoretical clear-sky irradiance model, and temperature forecasts provided by a meteorological service. The proposed method is validated on real data.

Introduction

A major challenge in the integration of renewable energy sources into the grid (Schiffer, Zonetti, Ortega, Stankovic, Sezi, & Raisch, 2016) is that power generation is intermittent, difficult to control, and strongly dependent on the variation of weather conditions. For this reason, forecasting of renewable distributed generation has become a fundamental requirement in order to reliably manage conventional power plant operation, grid balancing, real-time unit dispatching (Kim, Oh, Moore, & Ahn, 2016), demand constraints (Ishizaki et al., 2016), and energy market requirements. In this respect, renewable generation forecasts on different time horizons are of special interest to various players that operate in the active grid, in particular to Distribution System Operators (DSO) and Transmission System Operators (TSO) (see Albuyeh, 2009, Denholm and Margolis, 2007a, Denholm and Margolis, 2007b and references therein).

Concerning photovoltaic (PV) power generation (Coimbra, Kleissl, & Marquez, 2013), most contributions, focus on the problem of solar irradiance prediction (Inman et al., 2013, Kang and Tam, 2015, Perez et al., 2010). To tackle this problem, several approaches based on Artificial Neural Networks (ANNs) (Capizzi et al., 2012, Wu and Chan, 2011) or Support Vector Machines (Ragnacci, Pastorelli, Valigi, & Ricci, 2012) can be found in the literature. Alternatively, classical linear time series prediction methods are used in Bacher et al., 2009, Reikard, 2009, where the considered time series is typically the global horizontal irradiance (GHI) (Wong & Chow, 2001). GHI forecasts are typically used along with temperature forecasts in a simulation model of the PV plant (Patel, 2006) in order to calculate generated power predictions. In all cases, computing reliable forecasts from predicted meteorological variables hinges upon the availability of an accurate model of the plant, be it physical or estimated from data.

Unfortunately, in many common scenarios, neither a plant model, nor direct on-site measurements of solar irradiance and other meteorological variables (e.g., temperature) are available. This is always the case with a DSO dealing with hundreds or thousands of heterogeneous, independently owned and operated PV plants; in this case, the only available data consist of generated power measurements provided by smart meters, and of irradiance and temperature forecasts provided by a meteorological service. The problem of forecasting power generation in this case is addressed in Tao, Shanxu, and Changsong (2012) by means of a neural network and in Pepe et al., 2016, Pepe et al., 2017 using a parametric model. In these approaches, however, further information on the cloud cover index at the plant site is assumed to be available. In Bianchini et al., 2013a, Bianchini et al., 2013b, a heuristic method for the estimation of the parameters of the well-known PVUSA model (Dows & Gough, 1995) based on theoretical clear-sky irradiance is presented, while in Pepe, Bianchini, and Vicino (2018), a recursive procedure based on the clear-sky criteria proposed in Reno and Hansen (2016) is devised. However, the former approach does not allow for capturing possible parameter variations or seasonal drifts, and moreover both approaches require trial-and-error procedures in order to manually tune a number of algorithm parameters whose values may vary significantly according to the climate zone.

In this paper, a novel approach to the problem of estimating the parameters of the PVUSA model in the partial information case is presented. The only historical data used by the method consist of generated power, and temperature forecasts. Our approach is based on three tests to be performed on generated power data in order to detect portions of such data that were generated under clear-sky conditions. The information contained in such portions is then exploited in a recursive parameter estimation algorithm in combination with theoretical clear-sky irradiance data provided by a suitable model. The method proposed in this paper improves over (Bianchini et al., 2013a, Bianchini et al., 2013b, Pepe et al., 2018), since it is able to adapt to parameter variations and requires the tuning of a single threshold coefficient whose physical role is well defined.

The paper is structured as follows: in Section 2 the modeling tools are introduced; in Section 3 the proposed clear-sky detection tests are developed; the model estimation procedure is presented in Section 4. Experimental validation results are reported in Section 5, and conclusions are drawn in Section 6.

Section snippets

Preliminaries

A PV plant can be efficiently modeled using the PVUSA model (Dows & Gough, 1995), which expresses the instantaneous generated power as a function of irradiance and air temperature according to the equation: P=μ1I+μ2I2+μ3IT,where P, I, and T are the generated power (kW), irradiance (W/m2), and air temperature (∘C), respectively, and μ=[μ1μ2μ3]′ is the model parameter vector. It is important to notice that model (1) is linear in the parameters. For the purpose of this work, it is useful to

Clear-sky data detection

In this paper, the following key idea is exploited for the purpose of estimating the parameters of the PVUSA model (1) of a PV plant without resorting to on-site irradiance measurements. Given a time series composed of generated power measurements and temperature forecasts (or measurements, if available), we propose three tests to be performed on the data in order to detect portions of the power curve which have been generated under a clear-sky condition; this allows for fitting the parameters

Model estimation

According to the observations in the previous sections, we now introduce the proposed PVUSA plant model estimation method, which yields an on-line update of the parameter vector estimate μˆ by relying only on the information contained on a time series composed by measured power Pm and forecast (or measured) temperature T. The model estimation procedure is recursive, and combines tests T1, T2, T3 with a standard Recursive Least-Squares (RLS) algorithm using a dynamical time window. The following

Experimental results

Two experiments have been run to evaluate the performance of the proposed estimation algorithm. In the first one, both model estimation and validation have been conducted using measured data (power and temperature for estimation, irradiance and temperature for forecasting) in order to assess the performance of the estimation procedure net of errors due to inaccuracies of weather forecasts. In the second, meteorological predictions have been used both for model parameter fitting and generation

Conclusions

In this paper, an efficient technique for estimating a forecasting model of photovoltaic power generation from limited information has been proposed. The approach is based on a set of tests performed on power data combined with a recursive estimation framework. It only exploits the time series of generated power and forecasts of temperature, the latter obtained from a meteorological service. The procedure especially fits the typical scenario where the network operator has no access to on-site

Gianni Bianchini received the MS degree in electronic engineering from the University of Firenze, Italy, in 1997, and the PhD degree in control systems engineering from the University of Bologna, Italy, in 2001. In 2000, he was a visiting student at the Center for Control Engineering and Computation, University of California at Santa Barbara. In 2001–2002, he was a research associate at the University of Siena, and since 2003 he is an assistant professor of control systems at the same

References (33)

  • PerezR et al.

    Validation of short and medium term operational solar radiation forecasts in the US

    Solar Energy

    (2010)
  • ReikardG.

    Predicting solar radiation at high resolutions: A comparison of time series forecasts

    Solar Energy

    (2009)
  • RenoM.J. et al.

    Identification of periods of clear sky irradiance in time series of GHI measurements

    Renewable Energy

    (2016)
  • SchifferJ. et al.

    A survey on modeling of microgrids - From fundamental physics to phasors and voltage sources

    Automatica

    (2016)
  • WongL.T. et al.

    Solar radiation model

    Applied Energy

    (2001)
  • AlbuyehF.

    Grid of the future

    IEEE Power and Energy Magazine

    (2009)

    • Cited by (6)

    • Experimental validation of different PV power prediction models under Beni Mellal climate, implications for the energy nexus

      2022, Energy Nexus

    • A regression unsupervised incremental learning algorithm for solar irradiance prediction

      2021, Renewable Energy
      Citation Excerpt :

      They found that when the forecast horizon is between 1 and 3 h, frequency domain model has comparable results compared to ARIMA. Researchers in Ref. [28] presented a work to estimate photovoltaic generation using historical data such as generated power and past temperature forecasts without solar irradiance data. With the aid of theoretical clear-sky irradiance model, the work could produce good forecast despite lacking irradiance data.

    • Short-term forecasting photovoltaic solar power for home energy management systems

      2021, Inventions

    • Calculation of the real daily and monthly mean working temperature of a photovoltaic plant

      2021, Environmental Progress and Sustainable Energy

    • A Novel Prediction Error-Based Power Forecasting Scheme for Real PV System Using PVUSA Model: A Grey Box-Based Neural Network Approach

      2021, IEEE Access

    • Use of Synthetic Wind Power Time Series for Long-Term Voltage Stability Analysis

      2020, 2020 International Conference on Probabilistic Methods Applied to Power Systems, PMAPS 2020 - Proceedings

    Gianni Bianchini received the MS degree in electronic engineering from the University of Firenze, Italy, in 1997, and the PhD degree in control systems engineering from the University of Bologna, Italy, in 2001. In 2000, he was a visiting student at the Center for Control Engineering and Computation, University of California at Santa Barbara. In 2001–2002, he was a research associate at the University of Siena, and since 2003 he is an assistant professor of control systems at the same University. He is interested in energy systems, robust and nonlinear control, analysis and control of hybrid systems, aerospace applications, and haptics. He is associate editor of the European Journal of Control and past associate editor of IEEE Transactions on Circuits and Systems-II.

    Daniele Pepe received the BS degree cum laude in Automation Engineering in 2012, the MS degree cum laude in Computer and Automation Engineering in 2015, and the PhD in Information Engineering and Science in 2019, all from the University of Siena. He currently works in industry. His research interests include renewable energy forecasting and optimal energy management in buildings.

    Antonio Vicino is Professor Control Systems the Dept. of Information Engineering and Mathematical Sciences, University of Siena (Italy), where he covered the position of Head of the Department from 1996 to 1999 and Dean of the Engineering Faculty from 1999 to 2005. In 2000 he founded the Center for Complex Systems Studies of the University of Siena, where he covered the position of Director from 1999 to 2012. Since 2016 he is the Director of the PhD Program in Information Engineering and Science at the same university. Since May 2019 he is President of the Italian University Council. He is the author of 350 technical publications; co-author of the book Homogeneous Polynomial Forms for Robustness Analysis of Uncertain Systems (Springer-Verlag, 2009); co-editor of several books and guest editor of several special issues on robustness in identification and control. He has worked on stability analysis of nonlinear systems, time series analysis and prediction, robust control of uncertain systems. His present interests include identification of nonlinear systems, energy systems and smart grids, systems biology and applied system modeling.

    He has served as Associate Editor or Associate Editor at Large for several journals like IEEE Transactions on Automatic Control, IEEE Transactions on Circuits and Systems II and Automatica. He is Fellow of the IEEE and of the IFAC.

    ☆

    The material in this paper was not presented at any conference. This paper was recommended for publication in revised form by Associate Editor Angelo Alessandri under the direction of Editor Thomas Parisini

    View full text
    © 2019 Elsevier Ltd. All rights reserved.