Recalibration of recurrent neural networks for short-term wind power forecasting

Abstract This paper is focused on the day-ahead prediction of the onshore wind generation. This information is indeed published each day, ahead of the market clearing, by European Transmission System Operators (TSOs) to help market actors in their scheduling strategy. In that regard, our first objective is to improve the forecast performance by efficiently capturing the complex temporal dynamics of the wind power using recurrent neural networks. Practically, advanced architectures of Long Short Term Memory (LSTM) networks are implemented and compared. Secondly, in order to continuously refine the prediction tool, different techniques for recalibrating the model during its practical utilization are analyzed. This procedure consists in adjusting the parameters of the neural networks by taking advantage of the new information revealed over time, without the (time-consuming) need to retrain the model from scratch using the whole available dataset. Finally, the financial savings from the improvement of the forecast accuracy are estimated. Outcomes from the Belgian case study show that an optimal model recalibration can significantly improve forecast reliability, thereby decreasing the balancing costs of the system.

[1]  Peng Lu,et al.  Correlation-Constrained and Sparsity-Controlled Vector Autoregressive Model for Spatio-Temporal Wind Power Forecasting , 2018, IEEE Transactions on Power Systems.

[2]  Kit Po Wong,et al.  Probabilistic Forecasting of Wind Power Generation Using Extreme Learning Machine , 2014, IEEE Transactions on Power Systems.

[3]  Geoffrey E. Hinton,et al.  Deep Learning , 2015, Nature.

[4]  Ke Wang,et al.  A novel deep learning ensemble model with data denoising for short-term wind speed forecasting , 2020 .

[5]  Bart De Schutter,et al.  Forecasting spot electricity prices Deep learning approaches and empirical comparison of traditional algorithms , 2018 .

[6]  S. Santoso,et al.  Wind power forecasting and error analysis using the autoregressive moving average modeling , 2009, 2009 IEEE Power & Energy Society General Meeting.

[7]  Jie Zhang,et al.  Conditional aggregated probabilistic wind power forecasting based on spatio-temporal correlation , 2019, Applied Energy.

[8]  Anthony Papavasiliou,et al.  Dynamic dimensioning approach for operating reserves: Proof of concept in Belgium , 2019, Energy Policy.

[9]  Ma Li,et al.  Research on Short-Term Load Forecasting Using XGBoost Based on Similar Days , 2019, 2019 International Conference on Intelligent Transportation, Big Data & Smart City (ICITBS).

[10]  Arto Kaarna,et al.  Probabilistic Forecasting of Battery Energy Storage State-of-Charge under Primary Frequency Control , 2020, IEEE Journal on Selected Areas in Communications.

[11]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[12]  Haiyan Lu,et al.  A Novel Framework of Reservoir Computing for Deterministic and Probabilistic Wind Power Forecasting , 2020, IEEE Transactions on Sustainable Energy.

[13]  R. Buizza,et al.  Wind Power Density Forecasting Using Ensemble Predictions and Time Series Models , 2009, IEEE Transactions on Energy Conversion.

[14]  Jean-François Toubeau,et al.  Improved day-ahead predictions of load and renewable generation by optimally exploiting multi-scale dependencies , 2017, 2017 IEEE Innovative Smart Grid Technologies - Asia (ISGT-Asia).

[15]  Pierre Pinson,et al.  Impact of Public Aggregate Wind Forecasts on Electricity Market Outcomes , 2017, IEEE Transactions on Sustainable Energy.

[16]  David B. Stephenson,et al.  Best Practices for Postprocessing Ensemble Climate Forecasts. Part I: Selecting Appropriate Recalibration Methods , 2016 .

[17]  Jie Zhang,et al.  Multi-distribution ensemble probabilistic wind power forecasting , 2020, Renewable Energy.

[18]  Hui Liu,et al.  Smart wind speed forecasting approach using various boosting algorithms, big multi-step forecasting strategy , 2019, Renewable Energy.

[19]  Yi Wang,et al.  Combining Probability Density Forecasts for Power Electrical Loads , 2020, IEEE Transactions on Smart Grid.

[20]  Ming Yang,et al.  A Multi-Model Combination Approach for Probabilistic Wind Power Forecasting , 2017, IEEE Transactions on Sustainable Energy.

[21]  Jean-François Toubeau,et al.  Leveraging provision of frequency regulation services from wind generation by improving day-ahead predictions using LSTM neural networks , 2018, 2018 IEEE International Energy Conference (ENERGYCON).

[22]  F. Vallée,et al.  Medium-Term Multimarket Optimization for Virtual Power Plants: A Stochastic-Based Decision Environment , 2018, IEEE Transactions on Power Systems.

[23]  Jean-François Toubeau,et al.  Very-Short-Term Probabilistic Forecasting for a Risk-Aware Participation in the Single Price Imbalance Settlement , 2020, IEEE Transactions on Power Systems.

[24]  George Stavrakakis,et al.  Wind power forecasting using advanced neural networks models , 1996 .

[25]  Lieven Vandevelde,et al.  Droop Control as an Alternative Inertial Response Strategy for the Synthetic Inertia on Wind Turbines , 2016, IEEE Transactions on Power Systems.

[26]  M. Drinkwater,et al.  The advanced scatterometer (ASCAT) on the meteorological operational (MetOp) platform: A follow on for European wind scatterometers , 2002 .

[27]  Lieven Vandevelde,et al.  A Data-Driven Approach Using Deep Learning Time Series Prediction for Forecasting Power System Variables , 2019, 2019 IEEE 2nd International Conference on Renewable Energy and Power Engineering (REPE).

[28]  Jürgen Schmidhuber,et al.  Long Short-Term Memory , 1997, Neural Computation.

[29]  Daniel Kirschen,et al.  Combining Probabilistic Load Forecasts , 2018, IEEE Transactions on Smart Grid.

[30]  Jean-François Toubeau,et al.  Deep Learning-Based Multivariate Probabilistic Forecasting for Short-Term Scheduling in Power Markets , 2019, IEEE Transactions on Power Systems.