Atmospheric Boundary Layer Wind Profile Estimation Using Neural Networks Applied to Lidar Measurements

This paper introduces a new methodology for estimating the wind profile within the ABL (Atmospheric Boundary Layer) using a neural network and a single-point near-ground measurement. An important advantage of this solution when compared with others available in the literature is that it only requires near surface measurements for the prognosis once the neural network is trained. Another advantage is that it can be used to study the wind profile temporal evolution. This work uses data collected by a lidar sensor located at the Universidad de León (Spain). The neural network best configuration was determined using sensibility analyses. The result is a multilayer perceptron with three layers for each altitude: the input layer has six nodes for the last three measurements, the second has 128 nodes and the third consists of two nodes that provide u and v. The proposed method has better performance than traditional methods. The obtained wind profile information obtained is useful for multiple applications, such as preliminary calculations of the wind resource or CFD models.

[1]  Weihao Hu,et al.  High resolution wind speed forecasting based on wavelet decomposed phase space reconstruction and self-organizing map , 2019, Renewable Energy.

[2]  Ahmet Öztopal,et al.  Artificial neural network approach to spatial estimation of wind velocity data , 2006 .

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

[4]  Jesus Gonzalo de Grado,et al.  On the Development of a Digital Meteorological Model for Simulating Future Air Traffic Management Automation , 2011, 2011 IEEE 20th International Workshops on Enabling Technologies: Infrastructure for Collaborative Enterprises.

[5]  N. Srikanth,et al.  Suitability of power-law extrapolation for wind speed estimation on a tropical island , 2020, Journal of Wind Engineering and Industrial Aerodynamics.

[6]  A. García-Gutiérrez,et al.  An analysis and enhanced proposal of atmospheric boundary layer wind modelling techniques for automation of air traffic management , 2021 .

[7]  M. A. Mohandes,et al.  Wind Speed Extrapolation Using Machine Learning Methods and LiDAR Measurements , 2018, IEEE Access.

[8]  J. B. Duncan,et al.  Enhanced estimation of boundary layer advective properties to improve space‐to‐time conversion processes for wind energy applications , 2019, Wind Energy.

[9]  Na Zhang,et al.  A novel system based on neural networks with linear combination framework for wind speed forecasting , 2019, Energy Conversion and Management.

[10]  Abheejeet Mohapatra,et al.  Repeated wavelet transform based ARIMA model for very short-term wind speed forecasting , 2019, Renewable Energy.

[11]  Hacer Yumurtaci Aydogmus,et al.  The prediction of the wind speed at different heights by machine learning methods , 2016 .

[12]  Yongsheng Chen,et al.  Short-term wind speed and power forecasting using an ensemble of mixture density neural networks , 2016 .

[13]  Hiroto Kataoka,et al.  Applications and prospects of CFD for wind engineering fields , 2020 .

[14]  N. Bodini,et al.  The importance of round-robin validation when assessing machine-learning-based vertical extrapolation of wind speeds , 2020 .

[15]  Nikhil Ketkar,et al.  Introduction to Keras , 2017 .

[16]  Wei‐Chiang Hong,et al.  Chaos cloud quantum bat hybrid optimization algorithm , 2021 .

[17]  Diego Domínguez,et al.  A wind speed profile measurement method based on free bubble tracking in the lower atmosphere , 2013 .

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

[19]  Stefan Emeis,et al.  Surface-Based Remote Sensing of the Atmospheric Boundary Layer , 2010 .

[20]  Yi Zhou,et al.  On the Continuity of Rotation Representations in Neural Networks , 2018, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[21]  Ahsan Kareem,et al.  Emerging frontiers in wind engineering: Computing, stochastics, machine learning and beyond , 2020 .

[22]  Shuwen Zhang,et al.  A Review of Techniques for Diagnosing the Atmospheric Boundary Layer Height (ABLH) Using Aerosol Lidar Data , 2019, Remote. Sens..

[23]  Decreasing wind speed extrapolation error via domain-specific feature extraction and selection , 2020, Wind Energy Science.

[24]  Jae-Jin Kim,et al.  Wind speed estimation in urban areas based on the relationships between background wind speeds and morphological parameters , 2020 .

[25]  Jon Wieringa,et al.  Updating the Davenport roughness classification , 1992 .

[26]  Jin Wang,et al.  Improving WRF model turbine-height wind-speed forecasting using a surrogate- based automatic optimization method , 2019, Atmospheric Research.

[27]  Yuwei Wang,et al.  Short-term wind speed forecasting based on fast ensemble empirical mode decomposition, phase space reconstruction, sample entropy and improved back-propagation neural network , 2018 .

[28]  Yuan Yu,et al.  TensorFlow: A system for large-scale machine learning , 2016, OSDI.

[29]  Harish Gopalan,et al.  A coupled mesoscale–microscale framework for wind resource estimation and farm aerodynamics , 2014 .

[30]  Michael Y. Hu,et al.  Effect of data standardization on neural network training , 1996 .

[31]  Yunxiang Liu,et al.  Application of Neural Network to GNSS-R Wind Speed Retrieval , 2019, IEEE Transactions on Geoscience and Remote Sensing.

[32]  Shai Shalev-Shwartz,et al.  SGD Learns Over-parameterized Networks that Provably Generalize on Linearly Separable Data , 2017, ICLR.

[33]  R. Velo,et al.  Wind speed estimation using multilayer perceptron , 2014 .

[34]  Mohamed Mohandes,et al.  Short term wind speed estimation in Saudi Arabia , 2014 .

[35]  Harm de Vries,et al.  RMSProp and equilibrated adaptive learning rates for non-convex optimization. , 2015 .

[36]  Richard B. Alley,et al.  Advances in weather prediction , 2019, Science.

[37]  M. K. Soni,et al.  Artificial Neural Network-Based Peak Load Forecasting Using Conjugate Gradient Methods , 2002, IEEE Power Engineering Review.

[38]  Dave Winkler,et al.  Bayesian Regularization of Neural Networks , 2009, Artificial Neural Networks.

[39]  Priyanka Singh,et al.  Integration of new evolutionary approach with artificial neural network for solving short term load forecast problem , 2018 .

[40]  Dalibor Petković,et al.  Estimation of fractal representation of wind speed fluctuation by artificial neural network with different training algorothms , 2017 .

[42]  Hindayati Mustafidah,et al.  The most optimal performance of the Levenberg-Marquardt algorithm based on neurons in the hidden layer , 2019, Journal of Physics: Conference Series.

[43]  Diego Domínguez,et al.  Lighter-than-air particle velocimetry for wind speed profile measurement , 2014 .

[44]  Yoshihide Tominaga,et al.  Ten Questions Concerning Modeling of Near-Field Pollutant Dispersion in the Built Environment. , 2016 .

[45]  Marcello Ciampoli,et al.  Performance-based wind design of tall buildings , 2011 .

[46]  Lutz Prechelt,et al.  Automatic early stopping using cross validation: quantifying the criteria , 1998, Neural Networks.

[47]  Wei-Chiang Hong,et al.  Support Vector Regression Model Based on Empirical Mode Decomposition and Auto Regression for Electric Load Forecasting , 2013 .

[48]  Christoph W. Kent,et al.  Assessing methods to extrapolate the vertical wind-speed profile from surface observations in a city centre during strong winds , 2018 .