Review on probabilistic forecasting of wind power generation

By Yao Zhang, Jianxue Wang, Xifan Wang. 2014

The randomness and intermittence of wind resources is the biggest challenge in the integration of wind power into the power system. Accurate forecasting of wind power generation is an efficient tool to deal with such problem. Conventional wind power forecasting produces a value, or the conditional expectation of wind power output at a time point in the future. However, any prediction involves inherent uncertainty. In recent years, several probabilistic forecasting approaches have been reported in wind power forecasting studies. Compared to currently wide-used point forecasts, probabilistic forecasts could provide additional quantitative information on the uncertainty associated with wind power generation. For decision-makings in the uncertainty environment, probabilistic forecasts are optimal inputs. A review of state-of-the-art methods and new developments in wind power probabilistic forecasting is presented in this paper. Firstly, three different representations of wind power uncertainty are briefly introduced. Then, different forecasting methods are discussed. These methods are classified into three categories in terms of uncertainty representation, i.e. probabilistic forecasts (parametric and non-parametric), risk index forecasts and space-time scenario forecasts. Finally, requirements and the overall framework of the uncertainty forecasting evaluation are summarized. In addition, this article also describes current challenges and future developments associated with wind power probabilistic prediction.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Deep Forecast: Deep Learning-based Spatio-Temporal Forecasting (2017, background): “See (Zhang et al., 2014) for a comprehensive review of the state-of-the-art methods.”
Energy Forecasting: A Review and Outlook (2020, background): “A review on probabilistic wind power forecasting can be found in [12].”
High-resolution large-scale onshore wind energy assessments: A review of potential definitions, methodologies and future research needs (2021, unknown): “value given in the IEC61400-1 standard [16].”
Probabilistic energy forecasting: Global Energy Forecasting Competition 2014 and beyond (2016, background): “, 2014), and one to three papers for each of the tracks, including three for load forecasting (Hong& Fan, in this issue; Hong, Wilson, & Xie, 2014; Hyndman & Fan, 2010); one for price forecasting (Weron, 2014); two for wind forecasting (Pinson, 2013; Zhang et al., 2014); and two for solar forecastin”
Chance-Constrained Two-Stage Unit Commitment Under Uncertain Load and Wind Power Output Using Bilinear Benders Decomposition (2016, methodology): “The used probabilistic forecasting approach is non-parametric without any assumption of the density shape [41], [42].”
Deep Learning-Based Multivariate Probabilistic Forecasting for Short-Term Scheduling in Power Markets (2019, background): “In this way, different approaches for obtaining such uncertainty regions (or by extrapolation densities) can be found in the literature for respectively wind power [3]– [ 7 ] , PV generation [8], load [7], [9] and electricity prices [10], [11].”
Low-dimensional models in spatio-temporal wind speed forecasting (2015, unknown): “See [21] for a comprehensive review of the state-of-the-art methods.”
Recent advances in electricity price forecasting: A review of probabilistic forecasting (2016, background): “Note that there is one more commonly used attribute for probabilistic forecast evaluation, especially in the meteorological or wind power forecasting literature [18, 19].”
On-line quantile regression in the RKHS (Reproducing Kernel Hilbert Space) for operational probabilistic forecasting of wind power (2016, background): “A detailed review can be found in [10] 16 and [11].”