During the operation of wind turbines, there are a lot of abnormal data collected from wind power data, which cannot be used to truly evaluate the performance and operation state of wind turbines. According to the distribution characteristics of these abnormal data, an outlier delete method using Bayesian change point-quartile combined algorithm is proposed. Firstly, to identify and delete the abnormal data below wind power curve, based on the Bayesian theorem, change point position is identified. Finally, the quartile method is utilized to delete the upper sparse outliers. The proposed method is applied to 10-min field monitoring wind power data, root mean square error (RMSE), program running time and the rejection rate of field wind turbine data are used to measure model accuracy, the results show that compared with the least squares (LS) change point method, the method proposed in this paper is feasible and effective.
WuJShaoZYangS. A combined algorithm for data cleaning of wind power scatter diagram considering actual engineering characteristics. J Phys Conf Ser2020; 1639: 012044.
9.
ZhaoYYeLZhuQ. Characteristics and processing method of abnormal data cluster caused by wind curtailments in wind farms. Auto Electr Power Syst2014; 38(21): 39–46.
10.
ZhaoYYeLWangW, et al.Data-driven correction approach to refine power curve of wind farm under wind curtailment. IEEE Trans Sustain Energ2017; 9(1): 95–105.
11.
LiangTMengZCuiJ, et al.Health assessment of wind turbine based on laplacian eigenmaps. Energ Sourc A2020; 11: 1–15.
12.
LiangGSuYChenF, et al.Wind power curve data cleaning by image thresholding based on class uncertainty and shape dissimilarity. IEEE Trans Sustain Energ2020; 12(2): 1383–1393.
13.
LongHSangLWuZ, et al.Image-based abnormal data detection and cleaning algorithm via wind power curve. IEEE Trans Sustain Energ2019; 11(2): 938–946.
14.
WangZWangLHuangC. A fast abnormal data cleaning algorithm for performance evaluation of wind turbine. IEEE Trans Instrument Meas2020; 70: 1–12.
15.
LouJXuJLuH, et al.Wind turbine data-cleaning based on power curve. Auto Electr Power Syst2016; 40(10): 116–121.
16.
ZouTGaoYYiH, et al.Processing of wind power abnormal data based on Thompson tau-quartile and multi-point interpolation. Auto Electr Power Syst2020; 44(15): 156–162.
17.
WangYInfieldDStephenB, et al.Copula-based model for wind turbine power curve outlier rejection. Wind Energy2014; 17(11): 1677–1688.
18.
JingBQianZZareipourH, et al.Wind turbine power curve modelling with logistic functions based on quantile regression. Appl Sci2021; 11(7): 3048.
19.
XuKYanJZhangH, et al.Quantile based probabilistic wind turbine power curve model. Appl Energ2021; 296: 116913.
20.
CuiJYangJXingZ, et al.Multi-scenario wind power forecasting method based on single turbine optimal power curve fitting. Auto Electr Power Syst2015; 39(16): 23–28.
21.
ZhangQZhangXHouC. An automatic data cleaning and operating conditions classification method for wind turbines scada system. In: 17th International Computer Conference on Wavelet Active-Media Technology and Information Processing (ICCWAMTIP), Chengdu, China, 18-S21 December 2020, paper no. 437-443.
22.
ShenXFuXZhouC. A combined algorithm for cleaning abnormal data of wind turbine power curve based on change point grouping algorithm and quartile algorithm. IEEE Trans Sustain Energ2018; 10(1): 46–54.
23.
SuRZhangBShenC. A cleaning method for outliers in raw wind data based on the edge detection and variable-point variance. In: 2020 IEEE Sustainable Power and Energy Conference (iSPEC), Chengdu, China, 23-25 November, 2020, paper no. 1951-1957.
24.
ParkJLeeJOhK, et al.Development of a novel power curve monitoring method for wind turbines and its field tests. IEEE Trans Energ Convers2014; 29(1): 119–128.
25.
ChenJGuptaA. On change point detection and estimation. Commun Stat-simul C2001; 30(3): 665–697.
26.
FotopoulosSJandhyalaV. Maximum likelihood estimation of a change-point for exponentially distributed random variables. Stat Probabil Lett2001; 51(4): 423–429.
27.
DingJXiangYShenL, et al.Multiple change point analysis: fast implementation and strong consistency. IEEE Trans Signal Process2017; 65(17): 4495–4510.
28.
ChitsazanMFadaliMTrzynadlowskiA. Wind speed and wind direction forecasting using echo state network with nonlinear functions. Renew Energ2019; 131: 879–889.
