基于CFSFDP与LightGBM的风电机组异常状态预警研究

马良玉; 袁乃正

doi:10.19912/j.0254-0096.tynxb.2022-0001

PDF(2074 KB)

太阳能学报 ›› 2023, Vol. 44 ›› Issue (5) : 401-406. DOI: 10.19912/j.0254-0096.tynxb.2022-0001

基于CFSFDP与LightGBM的风电机组异常状态预警研究

马良玉, 袁乃正

作者信息 +

RESEARCH ON ABNORMAL CONDITION EARLY WARNING FOR WIND TURBINE BASED ON CFSFDP AND LIGHTGBM

Ma Liangyu, Yuan Naizheng

Author information +

文章历史 +

摘要

提出一种基于快速密度峰值聚类（CFSFDP）和LightGBM模型结合的风电机组异常状态监测方法。首先采用CFSFDP算法对风电机组监控与数据采集（SCADA）数据中的异常工况数据与噪声数据进行清洗;之后利用贝叶斯优化算法搜索LightGBM的最优超参数并建立风电机组正常运行工况预测模型。针对风速随机性的特点,利用时移滑动窗口方法构建异常状态识别指标,并结合核密度估计法计算指标阈值以实现异常工况预警。应用某风场的真实历史故障数据进行实验,结果表明LightGBM预警模型能在故障发生前对风电机组的异常工况进行及时正确的预警,验证了方法的有效性。

Abstract

A combination of clustering by fast search and find of density peaks（CFSFDP） and light gradient boosting machine（LightGBM） method is proposed for wind turbine abnormal condition monitoring. Firstly, the CFSFDP algorithm is employed to clean the abnormal condition data and noise data in the supervisory control and data acquisition system（SCADA）. Secondly, by using Bayesian optimization algorithm to search the optimal hyper-parameters of LightGBM, a LightGBM prediction model of wind turbine normal operation condition is established. Aiming at the randomness of wind speed, the time shift sliding window method is adopted to construct the abnormal state identification index and the kernel density estimation is employed to determine its threshold value. Finally, the actual historical faults data of a wind farm are used for experimental verification. The results show the LightGBM_model based early warning approach can warn the abnormal operation conditions of wind turbine timely before a fault occurs.

导出引用

马良玉, 袁乃正. 基于CFSFDP与LightGBM的风电机组异常状态预警研究[J]. 太阳能学报. 2023, 44(5): 401-406 https://doi.org/10.19912/j.0254-0096.tynxb.2022-0001

Ma Liangyu, Yuan Naizheng. RESEARCH ON ABNORMAL CONDITION EARLY WARNING FOR WIND TURBINE BASED ON CFSFDP AND LIGHTGBM[J]. Acta Energiae Solaris Sinica. 2023, 44(5): 401-406 https://doi.org/10.19912/j.0254-0096.tynxb.2022-0001

中图分类号： TK83

参考文献

[1] SHI Y L, LIU Y R, GAO X.Study of wind turbine fault diagnosis and early warning based on SCADA data[J]. IEEE access, 2021: 124600-124615.
[2] 金晓航, 许壮伟, 孙毅, 等. 基于SCADA数据分析和稀疏自编码神经网络的风电机组在线运行状态监测[J]. 太阳能学报, 2021, 42(6): 321-328.
JIN X H, XU Z W, SUN Y, et al.Online condition monitoring for wind turbines based on SCADA data analysis and sparse auto-encoder neural network[J]. Acta energiae solaris sinice, 2021, 42(6): 321-328.
[3] 沈小军, 付雪姣, 周冲成, 等. 风电机组风速-功率异常运行数据特征及清洗方法[J]. 电工技术学报, 2018, 33(14): 3353-3361.
SHEN X J, FU X J, ZHOU C C, et al.Characteristics of outliers in wind speed-power operation data of wind turbines and its cleaning method[J]. Transactions of China Electrotechnical Society, 2018, 33(14): 3353-3361.
[4] 赵永宁, 叶林, 朱倩雯. 风电场弃风异常数据簇的特征及处理方法[J]. 电力系统自动化, 2014, 38(21): 39-46.
ZHAO Y N, YE L, ZHU Q W.Characteristics and processing method of abnormal data clusters caused by wind curtailments in wind farms[J]. Automation of electric power systems, 2014, 38(21): 39-46.
[5] 马良玉, 孙佳明, 於世磊, 等. 基于DBSCAN和SDAE的风电机组异常工况预警研究[J]. 动力工程学报, 2021, 41(9): 786-793.
MA L Y, SUN J M, YU S L, et al.DBSCAN and SDAE-based abnormal condition early warning for a wind turbine unit[J]. Journal of Chinese Society of Power Engineering, 2021, 41(9): 786-793.
[6] 赵洪山, 闫西慧, 王桂兰, 等. 应用深度自编码网络和XG-Boost的风电机组发电机故障诊断[J]. 电力系统自动化, 2019, 43(1): 81-86.
ZHAO H S, YAN X H, WANG G L, et al.Fault diagnosis of wind turbine generator based on deep autoencoder network and XGBoost[J]. Automation of electric power systems, 2019, 43(1): 81-86.
[7] 崔恺, 许宜菲, 李雪松, 等. 基于广义回归神经网络的风电机组性能预测模型及状态预警[J]. 科学技术与工程, 2020, 20(32): 13220-13228.
CUI K, XU Y F, LI X S, et al.Wind turbine performance prediction model and early warning of abnormal condition based on GRNN[J]. Science technology and engineering, 2020, 20(32): 13220-13228.
[8] 睦浩淼. 基于支持向量机回归的风电机组运行状态异常检测方法研究[D]. 成都: 西南交通大学, 2018.
MU H M.Research on abnormality detection method of wind turbine operating state based on support vector machine regression[D]. Chengdu: Southwest Jiaotong University, 2018.
[9] RODRIGUEZ A, LAIO A.Clustering by fast search and find of density peaks[J]. Science, 2014, 344(6191): 1492-1496.
[10] GUO L K, QI M, FINLEY T, et al.LightGBM: a highly efficient gradient boosting decision tree[C]//Proceedings of the 2017 Advances in Neural Information Processing Systems, California: NIPS, 2017: 3146-3154.
[11] 颜诗旋, 朱平, 刘钊. 基于改进LightGBM模型的汽车故障预测方法研究[J]. 汽车工程, 2020, 42(6): 815-819.
YAN S X, ZHU P, LIU Z.Research on vehicle fault prediction scheme based on improved LightGBM model[J]. Automotive engineering, 2020, 42(6): 815-819.
[12] CHEN T, GUESTRIN C.Xgboost: a scalable tree boosting system[C]//Proceedings of the 22nd ACM Sigkdd International Conference on Knowledge Discovery and Data Mining, San Francisco, California, USA, 2016: 785-794.
[13] FRIEDMAN J H.Greedy function approximation: a gradient boosting machine[J]. Annals of statistics, 2001, 29(5): 1189-1232.
[14] 张帆, 刘德顺, 戴巨川, 等. 一种基于SCADA参数关系的风电机组运行状态识别方法[J]. 机械工程学报, 2019, 55(4): 1-9.
ZHANG F, LIU D S, DAI J C, et al.An operating condition recognition method of wind turbine based on SCADA parameter relations[J]. Journal of mechanical engineering, 2019, 55(4): 1-9.
[15] 杨楠, 周峥, 陈道君, 等. 基于非参数核密度估计的风功率波动性概率密度建模方法[J]. 太阳能学报, 2019, 40(7): 2028-2035.
YANG N, ZHOU Z, CHEN D J, et al.Research of modeling method based on non-parametric kernel density estimation of probability of wind power fluctuations[J]. Acta energiae solaris sinice, 2019, 40(7): 2028-2035.