Filters








7,361 Hits in 9.1 sec

Towards Better Wind Resource Modeling in Complex Terrain: A k-Nearest Neighbors Approach

Pedro Quiroga-Novoa, Gabriel Cuevas-Figueroa, José Luis Preciado, Rogier Floors, Alfredo Peña, Oliver Probst
2021 Energies  
Here, a novel approach for wind resource modeling is proposed, where a linearized flow model is combined with a machine learning approach based on the k-nearest neighbor (k-NN) method.  ...  The method was tested by performing cross-validations on a complex site using the measurements of five tall meteorological towers.  ...  All authors acknowledge the wind resource data from an undisclosed wind project developer, without which, this study would not have been possible.  ... 
doi:10.3390/en14144364 fatcat:phahayorpbd5zeaswc265ocqea

Wind Power Prediction with Machine Learning [chapter]

Nils André Treiber, Justin Heinermann, Oliver Kramer
2016 Studies in Computational Intelligence  
Hereby, we formulate the prediction task as regression problem and test different regression techniques such as linear regression, k-nearest neighbors and support vector regression.  ...  In our experiments, we analyze predictions for individual turbines as well as entire wind parks and show that a machine learning approach yields feasible results for short-term wind power prediction.  ...  Further, we thank the US National Renewable Energy Laboratory (NREL) for providing the wind data set.  ... 
doi:10.1007/978-3-319-31858-5_2 fatcat:c2wyjdtx7vefzkfbq27xwhmyiq

Improved Weighted k-Nearest Neighbor Based on PSO for Wind Power System State Recognition

Chun-Yao Lee, Kuan-Yu Huang, Yi-Xing Shen, Yao-Chen Lee
2020 Energies  
In this paper, we propose using particle swarm optimization (PSO) which can improve weighted k-nearest neighbors (PWKNN) to diagnose the failure of a wind power system.  ...  A comparison with other traditional classifiers, i.e., backpropagation neural network (BPNN), k-nearest neighbor (k-NN), and radial basis function network (RBFN) shows that PWKNN has a higher classification  ...  The PSO optimizes weights and k value and estimates the predictive classification accuracy of PWKNN with leave one out cross-validation (LOOCV). In the LOOCV, each sample data is a class.  ... 
doi:10.3390/en13205520 fatcat:hdz77e6jeffhjosqstxwidupea

A Machine Learning-Based Gradient Boosting Regression Approach for Wind Power Production Forecasting: A Step towards Smart Grid Environments

Upma Singh, Mohammad Rizwan, Muhannad Alaraj, Ibrahim Alsaidan
2021 Energies  
Scatter curves depicting relationships between the wind speed and the produced turbine power are plotted for all of the methods and the predicted average wind power is compared with the real average power  ...  In the present study, we propose and compare five optimized robust regression machine learning methods, namely, random forest, gradient boosting machine (GBM), k-nearest neighbor (kNN), decision-tree,  ...  Data Availability Statement: The scrutiny of data and forecasting were accomplished with the openly available dataset that has been collected from the SCADA system for wind turbines in the northwestern  ... 
doi:10.3390/en14165196 fatcat:6xgb7bb5uvdpvabfzchctilzaq

Structural monitoring for lifetime extension of offshore wind monopiles: Can strain measurements at one level tell us everything?

Lisa Ziegler, Ursula Smolka, Nicolai Cosack, Michael Muskulus
2017 Wind Energy Science Discussions  
Damage equivalent loads at unmeasured locations are predicted from the simulation results with a k-nearest neighbor regression algorithm.  ...  Results show that damage can be predicted with an error of 1–3 % if this is done conditional on mean wind speed, which is very promising.  ...  Predict desired M-DEL as mean or a weighted mean of the simulated M-DEL values corresponding to the nearest neighbor simulated T-DELs.  ... 
doi:10.5194/wes-2017-21 fatcat:ilpcxdao4feyberazfwooqhp24

Brief communication: Structural monitoring for lifetime extension of offshore wind monopiles: can strain measurements at one level tell us everything?

Lisa Ziegler, Ursula Smolka, Nicolai Cosack, Michael Muskulus
2017 Wind Energy Science  
Damage equivalent loads at unmeasured locations are predicted from the simulation results with a <i>k</i>-nearest neighbor regression algorithm.  ...  Results show that damage can be predicted with an error of 1–3<span class="thinspace"></span>% if this is done conditional on mean wind speed, which is very promising.  ...  Predict desired M-DEL as mean or a weighted mean of the simulated M-DEL values corresponding to the nearest neighbor simulated T-DELs.  ... 
doi:10.5194/wes-2-469-2017 fatcat:ljewsiquvjheda2clx3gcy2wdu

Extended Coopetitive Soft Gating Ensemble [article]

Stephan Deist, Jens Schreiber, Maarten Bieshaar, Bernhard Sick
2020 arXiv   pre-print
The extended CSGE (XCSGE as we term it), is used to predict the power generation on both wind- and solar farms.  ...  The XCSGE reaches an improvement of the prediction performance of up to 11% for wind power forecasting and 30% for solar power forecasting compared to the worst performing model.  ...  We use a k-nearest neighbor regressor as the machine learning model M C. Wind farms In this chapter, the XCSGE is applied to wind farm datasets.  ... 
arXiv:2004.14026v1 fatcat:xw7a3flkrjbehagpby56emkdpm

Analysis of cedar pollen time series: no evidence of low-dimensional chaotic behavior

J.-J. Delaunay, R. Konishi, C. Seymour
2005 International journal of biometeorology  
We believe that our conclusion can be generalized to pollen series from other wind-pollinated plant species, as wind speed, the forcing parameter of the pollen emission and transport, is best described  ...  We showed that the choice of test for detection of deterministic chaos in pollen series is difficult because pollen series exhibit 1=f power spectra.  ...  The predictions are obtained using a nonlinear prediction scheme such as the k-nearest neighbor scheme. We used the local constant approximation (zeroth order) of the k-nearest neighbor scheme.  ... 
doi:10.1007/s00484-005-0004-9 pmid:16208500 fatcat:7idsleyrejcs3jwmpfa3sll66u

A Stochastic Approach for Prediction of Partially Measured Concentrations of Benzo[a]pyrene in the Ambient Air in Korea

Yongku Kim, Young-Kyo Seo, Kyung-Min Baek, Min-Ji Kim, Sung-Ok Baek
2016 Asian Journal of Atmospheric Environment  
The proposed approach constructs a nearest-neighbor structure by incorporating the linkage between BaP and meteorology and meteorological effects.  ...  This approach is adopted in order to predict unobserved BaP concentrations based on observed (or forecasted) meteorological conditions, including temperature, precipitation, wind speed, and air quality  ...  The weight function gives more weight to the nearest neighbors and less to the farthest neighbors.  ... 
doi:10.5572/ajae.2016.10.4.197 fatcat:cs3owyububaclgvjt6vjoexzki

Investigation of short-range cedar pollen forecasting

J.-J. Delaunay, C. Seymour, V. Fouillet
2004 Physical Review E  
The nearest-neighbor method using local constant prediction applied to hourly pollen forecasting with a 1-h lead time was effective for small to medium pollen variations, but failed to reproduce large  ...  Pollen forecasting is of increasing interest as a way to help the general public avoid contact with allergyinducing pollen.  ...  We varied the embedding dimension m and the number of nearest neighbors k of the prediction scheme, tried regular zeroth-order and distance weighted zeroth-order approaches, and applied a nonlinear filtering  ... 
doi:10.1103/physreve.70.066214 pmid:15697491 fatcat:yemgu2eravgmng4ehmiak7fc7e

Assessment of Equipment Operation State with Improved Random Forest

Na Yang, She Liu, Jie Liu, Changjie Li, Ashwani K. Gupta
2021 International Journal of Rotating Machinery  
After the decision trees with different classification capabilities are weighted, an IRF model is established.  ...  The method has a good application prospect in the state assessment of wind power equipment.  ...  That is, k (usually 5)-nearest neighbor samples are found for each sample in the minority data sets.  ... 
doi:10.1155/2021/8813443 fatcat:jrxoni5dcndalmynsma2kakq3i

Synchronization of oscillators with long-range power law interactions

Debanjan Chowdhury, M. C. Cross
2010 Physical Review E  
We present analytical calculations and numerical simulations for the synchronization of oscillators interacting via a long range power law interaction on a one dimensional lattice.  ...  We have identified the critical value of the power law exponent α_c across which a transition from a synchronized to an unsynchronized state takes place for a sufficiently strong but finite coupling strength  ...  model ͑K ij = K / N͒ and the short range model, where only the nearest neighbors interact with each other ͑K ij = K for nearest neighbors, zero otherwise͒ have been studied in great detail ͓5͔.  ... 
doi:10.1103/physreve.82.016205 pmid:20866705 fatcat:v2qmcjbq5jbpzhnzkuwglfm2my

Fault Diagnosis for Wind Turbines Based on ReliefF and eXtreme Gradient Boosting

Zidong Wu, Xiaoli Wang, Baochen Jiang
2020 Applied Sciences  
According to the SCADA history data and the wind turbines fault record, the ReliefF algorithm is used to select feature parameters that are highly correlated with common faults.  ...  First of all, we use the historical data records as the input, and use the ReliefF algorithm to select the SCADA system observation features with high correlation with the fault classification, then use  ...  where W(T) i represents the weight of the t-th feature in the i-th iteration, and the initial value is 0, H j represents k nearest neighbor samples of the same kind as R, M j (C) represents k nearest  ... 
doi:10.3390/app10093258 fatcat:6ruhjptrobfnrkkcylouzmemha

Predicting Fine Particulate Matter (PM2.5) in the Greater London Area: An Ensemble Approach using Machine Learning Methods

Mahdieh Danesh Yazdi, Zheng Kuang, Konstantina Dimakopoulou, Benjamin Barratt, Esra Suel, Heresh Amini, Alexei Lyapustin, Klea Katsouyanni, Joel Schwartz
2020 Remote Sensing  
The ensemble included predictions from three different machine learners: a random forest (RF), a gradient boosting machine (GBM), and a k-nearest neighbor (KNN) approach.  ...  Our ensemble model performed very well, with a ten-fold cross-validated R2 of 0.828. Of the three machine learners, the random forest outperformed the GBM and KNN.  ...  For the k-nearest neighbor, we found the optimal value of "k" using cross-validation.  ... 
doi:10.3390/rs12060914 fatcat:62taz4xpgjb7npth4kyvnivmcm

Wind Speed Forecasting in China: A Review

Huiru Zhao
2015 Science Journal of Energy Engineering  
China's wind power has developed rapidly in the past few years, the large-scale penetration of which will bring big influence on power systems.  ...  The wind speed forecasting research is quite important because it can alleviate the negative impacts. This paper reviews the current wind speed forecasting techniques in China.  ...  Then, wind speed chaotic prediction model of optimal neighborhood was proposed which gives overall consideration to the nearest neighbors' weights and generalized degrees of freedom, also an improved criterion  ... 
doi:10.11648/j.sjee.s.2015030401.13 fatcat:dpbqaw57o5c6pm7ozgnv7zwppe
« Previous Showing results 1 — 15 out of 7,361 results