针对传统自组织核回归（AAKR）模型所选记忆矩阵冗余度较高、无法根据在线数据实时更新、计算相似度时未考虑特征参数权值不一的问题,提出一种基于动态矩阵与特征相似度的自组织核回归（DM-FS-AAKR）风电机组状态监测方法。首先基于样本间距离对原始数据集去冗余以降低运算复杂度,形成待选数据集;其次基于k-最近邻算法选取最符合当前运行条件的历史数据构建动态矩阵;为克服相似度计算时不良参数的偏差污染,提出一种特征相似度计算方法为不同参数分配相应权值进一步提高预测精度;最后以河北某风电场SCADA数据为例,对机组故障停机前工况进行验证实验。结果表明,相比于传统AAKR模型,所提算法平均绝对误差降低约15.6%,故障预警时能够提前35天实现预警,具有较高精度和实时性。
Abstract
A dynamic matrix and feature similarity-based Auto Associative Kernel Regression （DM-FS-AAKR） for condition monitoring of wind turbine is proposed to address the problems of high redundancy in the memory matrix selected by the traditional auto associative kernel regression model, inability to update in real-time based on online data, and failure to consider inconsistent feature parameter weights when calculating similarity. Firstly, based on the distance between samples, the original dataset is de-redundant to reduce computational complexity and form a dataset to be selected; Secondly, based on the k-nearest neighbor algorithm, the dynamic matrix is constructed by selecting the historical data that best meets the current operating conditions. To overcome the bias pollution of bad parameters during similarity calculation, a feature similarity calculation method is proposed to assign corresponding weights to different parameters to further improve prediction accuracy. Finally, taking the SCADA data of a wind farm in Hebei as an example, simulation verification experiments were conducted on the operating conditions before the unit malfunctions and shuts down. The results show that compared to the traditional AAKR model, the proposed algorithm reduces the average absolute error by about 15.6%, and the fault warning can be achieved 35 days in advance, with high accuracy and real-time performance.
关键词
齿轮箱 /
风电机组 /
状态监测 /
自组织核回归 /
动态矩阵 /
特征相似度
Key words
gearbox /
wind turbines /
condition monitoring /
auto associative kernel regression（AAKR） /
dynamic matrix /
feature similarity
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 刘杰, 曹静, 赵昕. 基于OOB-GWO-SVR的风电机组齿轮箱故障预警[J]. 电子测量与仪器学报, 2022, 36(12): 97-105.
LIU J, CAO J, ZHAO X.Wind turbine gearbox fault warning based on OOB-GWO-SVR[J]. Journal of electronic measurement and instrumentation, 2022, 36(12): 97-105.
[2] 罗志宏, 刘长良, 刘帅. 基于近邻元分析的风电机组状态监测特征选择方法[J]. 华北电力大学学报(自然科学版), 2024, 51(3): 134-142.
LUO Z H, LIU C L, LIU S.A feature selection method based on neighborhood component analysis for wind turbine condition monitoring[J]. Journal of North China Electric Power University (natural science edition), 2024, 51(3): 134-142.
[3] 任建亭, 汤宝平, 雍彬, 等. 基于深度变分自编码网络融合SCADA数据的风电齿轮箱故障预警[J]. 太阳能学报, 2021, 42(4): 403-408.
REN J T, TANG B P, YONG B, et al.Wind turbine gearbox fault warning based on depth variational autoencoders network fusion SCADA data[J]. Acta energiae solaris sinica, 2021, 42(4): 403-408.
[4] 王梓齐, 张书瑶, 刘长良. 基于增量式相对熵的风电机组实时状态监测[J]. 电子测量与仪器学报, 2020, 34(12): 125-132.
WANG Z Q, ZHANG S Y, LIU C L.Real-time condition monitoring of wind turbine based on incremental relative entropy[J]. Journal of electronic measurement and instrumentation, 2020, 34(12): 125-132.
[5] PARK J, KIM C, DINH M C, et al.Design of a condition monitoring system for wind turbines[J]. Energies, 2022, 15(2): 464.
[6] 蒋佳炜, 胡以怀, 李方玉, 等. 基于AAKR模型的船用低速柴油机状态监测方法[J]. 舰船科学技术, 2021, 43(2): 158-162.
JIANG J W, HU Y H, LI F Y, et al.Ship low-speed diesel engine condition monitoring method based on AAKR model[J]. Ship science and technology, 2021, 43(2): 158-162.
[7] YU H, LI H R.Pump remaining useful life prediction based on multi-source fusion and monotonicity-constrained particle filtering[J]. Mechanical systems and signal processing, 2022, 170: 108851.
[8] 褚景春, 郭鹏, 解加盈. 自组织核回归风电机组功率曲线建模与应用研究[J]. 太阳能学报, 2021, 42(7): 372-377.
CHU J C, GUO P, XIE J Y.Wind turbine power curve modeling and application based on AAKR method[J]. Acta energiae solaris sinica, 2021, 42(7): 372-377.
[9] 刘长良, 张书瑶, 王梓齐. 基于改进KNN回归算法的风电机组齿轮箱状态监测[J]. 中国测试, 2021, 47(1): 153-159.
LIU C L, ZHANG S Y, WANG Z Q.Condition monitoring of wind turbine gearbox based on improved KNN regression algorithm[J]. China measurement & test, 2021, 47(1): 153-159.
[10] AN S H, HEO G, CHANG S H.Detection of process anomalies using an improved statistical learning framework[J]. Expert systems with applications, 2011, 38(3): 1356-1363.
[11] WANG Z Q, LIU C L, YAN F.Condition monitoring of wind turbine based on incremental learning and multivariate state estimation technique[J]. Renewable energy, 2022, 184: 343-360.
[12] 刘双白, 朱龙飞, 仇晓智, 等. 基于加权AAKR算法的发电设备状态预警技术研究[J]. 热能动力工程, 2020, 35(7): 235-241.
LIU S B, ZHU L F, QIU X Z, et al.Study of power equipment condition early warning technology based on weighted AAKR algorithm[J]. Journal of engineering for thermal energy and power, 2020, 35(7): 235-241.
[13] YU W K, ZHAO C H, HUANG B.MoniNet with concurrent analytics of temporal and spatial information for fault detection in industrial processes[J]. IEEE transactions on cybernetics, 2022, 52(8): 8340-8351.
[14] 刘纪平, 梁恩婕, 徐胜华, 等. 顾及样本优化选择的多核支持向量机滑坡灾害易发性分析评价[J]. 测绘学报, 2022, 51(10): 2034-2045.
LIU J P, LIANG E J, XU S H, et al.Multi-kernel support vector machine considering sample optimization selection for analysis and evaluation of landslide disaster susceptibility[J]. Acta geodaetica et cartographica sinica, 2022, 51(10): 2034-2045.
基金
中央高校基本科研业务费专项资金（2020JG006; 2023JG005; 2023JC010）; 河北省高等学校科学技术研究项目（CXY2023001）
PDF(3077 KB)
