Assessment of critical parameters for artificial neural networks based short-term wind generation forecasting

Abstract Participation of wind energy in the generation portfolio of power systems is increasing, making it more challenging for system operators to adequately maintain system security. It therefore becomes increasingly crucial to accurately predict the wind generation. This work investigates how different parameters influence the performance of forecasting algorithms. Firstly, this work analyzes the combined influence of the input data, batch size, number of neurons and hidden layers, and the training data on the forecast accuracy across forecast horizons of 5, 15, 30 and 60 min. It was found that increasing look ahead times require among others more hidden layers and lower batch sizes. Next, the optimizer and loss function leading to the most accurate forecasts were identified. It was concluded that the Adadelta optimizer and Mean Absolute Error loss function consistently result in the best performing forecasting algorithm. Finally, it was investigated if the most accurate optimizer-loss function combination is influenced by the choice of the performance metric. Whereas the Adadelta-Mean Absolute Error pair remains the most accurate combination irrespective of the evaluation metric, a strong relation was observed between the Root Mean Square Error performance metric and Mean Square Error loss function. Analyses were performed on 12 wind farms.

[1]  Bri-Mathias Hodge,et al.  Overview and Meteorological Validation of the Wind Integration National Dataset toolkit , 2015 .

[2]  Jing Shi,et al.  On comparing three artificial neural networks for wind speed forecasting , 2010 .

[3]  Jian Sun,et al.  Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification , 2015, 2015 IEEE International Conference on Computer Vision (ICCV).

[4]  Yoshua Bengio,et al.  Deep Sparse Rectifier Neural Networks , 2011, AISTATS.

[5]  Yoram Singer,et al.  Adaptive Subgradient Methods for Online Learning and Stochastic Optimization , 2011, J. Mach. Learn. Res..

[6]  J. Mahmoudi,et al.  Short and mid-term wind power plants forecasting with ANN , 2012, 2012 Second Iranian Conference on Renewable Energy and Distributed Generation.

[7]  H. Robbins.  A Stochastic Approximation Method , 1951 .

[8]  S. Agatonovic-Kustrin,et al.  Basic concepts of artificial neural network (ANN) modeling and its application in pharmaceutical research. , 2000, Journal of pharmaceutical and biomedical analysis.

[9]  Abdul Mouez Khatab.  Performance Analysis of Operating Wind Farms , 2017 .

[10]  Josep M. Guerrero,et al.  Hybrid machine intelligent SVR variants for wind forecasting and ramp events , 2019, Renewable and Sustainable Energy Reviews.

[11]  Geoffrey E. Hinton,et al.  Rectified Linear Units Improve Restricted Boltzmann Machines , 2010, ICML.

[12]  Hao Hu,et al.  Optimization of an Artificial Neural Network System for the Prediction of Failure Analysis Success , 2019, Microelectronics Reliability.

[13]  Bri-Mathias Hodge,et al.  Ramp forecasting performance from improved short-term wind power forecasting over multiple spatial and temporal scales , 2017 .

[14]  Rob J Hyndman,et al.  Another look at measures of forecast accuracy , 2006 .

[15]  B. Hodge,et al.  The value of day-ahead solar power forecasting improvement , 2016 .

[16]  Matthew D. Zeiler.  ADADELTA: An Adaptive Learning Rate Method , 2012, ArXiv.

[17]  Léon Bottou,et al.  Stochastic Gradient Descent Tricks , 2012, Neural Networks: Tricks of the Trade.

[18]  Honglak Lee,et al.  Understanding and Improving Convolutional Neural Networks via Concatenated Rectified Linear Units , 2016, ICML.

[19]  Bri-Mathias Hodge,et al.  The Wind Integration National Dataset (WIND) Toolkit , 2015 .

[20]  Yoshua Bengio,et al.  Understanding the difficulty of training deep feedforward neural networks , 2010, AISTATS.

[21]  T. Bruns,et al.  Differences in Wind Forecast Accuracy in the German North and Baltic Seas , 2014 .

[22]  Bri-Mathias Hodge,et al.  The value of improved wind power forecasting: Grid flexibility quantification, ramp capability analysis, and impacts of electricity market operation timescales , 2016 .

[23]  Bri-Mathias Hodge,et al.  Validation of Power Output for the WIND Toolkit , 2014 .

[24]  Nilay Shah,et al.  Machine-learning methods for integrated renewable power generation: A comparative study of artificial neural networks, support vector regression, and Gaussian Process Regression , 2019, Renewable and Sustainable Energy Reviews.

[25]  Dereje Shiferaw,et al.  Short-Term Wind Power Forecasting Using Artificial Neural Networks for Resource Scheduling in Microgrids , 2016 .

[26]  Pierre Pinson,et al.  Standardizing the Performance Evaluation of Short-Term Wind Power Prediction Models , 2005 .

[27]  Hufang Yang,et al.  An innovative hybrid system for wind speed forecasting based on fuzzy preprocessing scheme and multi-objective optimization , 2019, Energy.

[28]  Ulrich Focken,et al.  Short-term prediction of the aggregated power output of wind farms—a statistical analysis of the reduction of the prediction error by spatial smoothing effects , 2002 .

[29]  P.J. Zucatelli,et al.  Short-term wind speed forecasting in Uruguay using computational intelligence , 2019, Heliyon.

[30]  Minping Jia,et al.  Wear indicator construction of rolling bearings based on multi-channel deep convolutional neural network with exponentially decaying learning rate , 2019, Measurement.

[31]  Paul Leahy,et al.  Quantifying the value of improved wind energy forecasts in a pool-based electricity market , 2015 .

[32]  K. A. Folly,et al.  Wind Power Forecasting , 2018 .

[33]  P. Pinson,et al.  What performance can be expected by short-term wind power prediction models depending on site characteristics? , 2004 .

[34]  Han Zhang,et al.  Research on a learning rate with energy index in deep learning , 2019, Neural Networks.

[35]  Sebastian Ruder,et al.  An overview of gradient descent optimization algorithms , 2016, Vestnik komp'iuternykh i informatsionnykh tekhnologii.

[36]  Ronald Davis,et al.  Neural networks and deep learning , 2017 .

[37]  Jonathan Collins,et al.  Short Term Forecasting for Utility-Scale Wind Farms — The Power Model Challenge , 2009, 2009 CIGRE/IEEE PES Joint Symposium Integration of Wide-Scale Renewable Resources Into the Power Delivery System.

[38]  Timothy Dozat,et al.  Incorporating Nesterov Momentum into Adam , 2016 .

[39]  Mohammad Monfared,et al.  A new strategy for wind speed forecasting using artificial intelligent methods , 2009 .

[40]  Qing Cao,et al.  Forecasting wind speed with recurrent neural networks , 2012, Eur. J. Oper. Res..

[41]  William D'haeseleer,et al.  Optimization and allocation of spinning reserves in a low-carbon framework , 2016, 2016 IEEE Power and Energy Society General Meeting (PESGM).

[42]  T. Chai,et al.  Root mean square error (RMSE) or mean absolute error (MAE)? – Arguments against avoiding RMSE in the literature , 2014 .

[43]  Marc A. Rosen,et al.  Short-term wind speed forecasting using artificial neural networks for Tehran, Iran , 2016, International Journal of Energy and Environmental Engineering.

[44]  Erik Ela,et al.  ERCOT Event on February 26, 2008: Lessons Learned , 2008 .

[45]  Xing Liu,et al.  An improved multi-branch residual network based on random multiplier and adaptive cosine learning rate method , 2019, J. Vis. Commun. Image Represent..

[46]  J.B. Theocharis,et al.  Long-term wind speed and power forecasting using local recurrent neural network models , 2006, IEEE Transactions on Energy Conversion.

[47]  Caroline Draxl,et al.  Guide to Using the WIND Toolkit Validation Code , 2014 .

[48]  Jianzhou Wang,et al.  A novel hybrid system based on multi-objective optimization for wind speed forecasting , 2020 .

[49]  Kaushik Das,et al.  Adequacy of operating reserves for power systems in future european wind power scenarios , 2015, 2015 IEEE Power & Energy Society General Meeting.

[50]  Jari Miettinen,et al.  Wind power forecasting accuracy and uncertainty in Finland , 2013 .

[51]  Joao P. S. Catalao,et al.  Short-term wind power forecasting in Portugal by neural networks and wavelet transform , 2011 .

[52]  Yoshua Bengio,et al.  Practical Recommendations for Gradient-Based Training of Deep Architectures , 2012, Neural Networks: Tricks of the Trade.

[53]  Puneet Kumar,et al.  Solar radiation forecasting using artificial neural network with different meteorological variables , 2016 .

[54]  I. González-Aparicio,et al.  Impact of wind power uncertainty forecasting on the market integration of wind energy in Spain , 2015 .

[55]  W. Rivera,et al.  Short term wind speed forecasting in La Venta, Oaxaca, México, using artificial neural networks , 2009 .

[56]  Jianzhou Wang,et al.  Short-term wind speed forecasting using a hybrid model , 2017 .

[57]  Muhammad Khalid,et al.  A review on the selected applications of forecasting models in renewable power systems , 2019, Renewable and Sustainable Energy Reviews.

[58]  Jianguo Zhou,et al.  Short-Term Wind Power Forecasting: A New Hybrid Model Combined Extreme-Point Symmetric Mode Decomposition, Extreme Learning Machine and Particle Swarm Optimization , 2018, Sustainability.

[59]  Jinchun Song,et al.  Multi-Objective Optimization Design through Machine Learning for Drop-on-Demand Bioprinting , 2019, Engineering.

[60]  Muhammad Khalid,et al.  Multi-step Ahead Wind Forecasting Using Nonlinear Autoregressive Neural Networks , 2017 .

[61]  Georg Heigold,et al.  An empirical study of learning rates in deep neural networks for speech recognition , 2013, 2013 IEEE International Conference on Acoustics, Speech and Signal Processing.

[62]  J. Zack,et al.  Application of bias corrections to improve hub-height ensemble wind forecasts over the Tehachapi Wind Resource Area , 2019, Renewable Energy.