针对风力发电生产过程中齿轮箱常发故障,结合工业过程多变量、高维度、非线性、数据关联度高、异常数据缓慢劣化等特点,采用滑动窗口-核熵主元分析法（MW-KECA）对齿轮箱劣变过程中机组相关振动数据进行监控,并对齿轮箱故障进行诊断。改善样本适应性,同时针对算法中核参数选择偶然性和随机性问题,引入鲸鱼优化算法进行改进。通过齿轮箱数据集进行仿真研究,并在滚动轴承实验台的故障诊断中实现不同故障类型的准确识别验证,通过对比分析,采用MW-KECA和鲸鱼算法识别结果可更高效和准确地预测故障时间,减小误报率。
Abstract
A Moving Window-Kernel Entropy Component Analysis （MW-KECA） approach was used to monitor the vibration data related to the units during the gearbox degradation process and diagnose gearbox faults in response to the frequent failures in gearboxes during the production process of wind power generation.This was done while considering the characteristics of industrial processes, such as multiple variables, high dimensionality, non-linearity, high data correlation, and slow deterioration of abnormal data.This method improves sample overfitting.Additionally,the Whale Optimization Algorithm （WOA） was introduced to address the issues of stochasticity and randomness in kernel parameter selection inherent in the algorithm.Simulation studies were conducted using gearbox datasets, and accurate identification and verification of different fault types were achieved in fault diagnosis on a rolling bearing test rig.Comparative analyses reveals that the adoption of MW-KECA combined with WOA results in more efficient and accurate fault time prediction, with a reduced false alarm rate.
关键词
风电机组 /
齿轮箱 /
故障诊断 /
滑动窗口 /
主成分分析 /
预测分析
Key words
wind turbines /
gearbox /
fault diagnosis /
sliding window /
principal component analysis /
predictive analysis
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 宋中越, 马姣姣, 甄冬, 等. 风力发电机齿轮箱高故障率原因的研究及改进措施[J]. 机床与液压, 2018, 46(1): 173-177.
SONG Z Y, MA J J, ZHEN D, et al.Research of reason of high failure rate and improvement measures about gearbox of wind turbine[J]. Machine tool & hydraulics, 2018, 46(1): 173-177.
[2] SCHÖLKOPF B, SMOLA A, MÜLLER K R. Nonlinear component analysis as a kernel eigenvalue problem[J]. Neural computation, 1998, 10(5): 1299-1319.
[3] JIANG Q C, YAN X F, LV Z M, et al.Fault detection in nonlinear chemical processes based on kernel entropy component analysis and angular structure[J]. Korean journal of chemical engineering, 2013, 30(6): 1181-1186.
[4] 郭金玉, 李涛, 李元. 基于滑动窗口KECA-SVM的非线性过程故障检测[J]. 沈阳大学学报(自然科学版), 2022, 34(1): 37-44.
GUO J Y, LI T, LI Y.Fault detection of nonlinear process based on moving window KECA-SVM[J]. Journal of Shenyang University (natural science), 2022, 34(1): 37-44.
[5] 许昕, 韩慧苗, 潘宏侠, 等. 基于多特征提取和KECA柴油机关键部件故障识别[J]. 电子测量技术, 2021, 44(19): 63-68.
XU X, HAN H M, PAN H X, et al.Fault identification of key components of diesel engine based on multi feature extraction and KECA[J]. Electronic measurement technology, 2021, 44(19): 63-68.
[6] 祝志超, 吴定会, 岳远昌. 基于监督核熵成分分析的发动机磨损故障诊断[J]. 系统仿真学报, 2022, 34(1): 45-52.
ZHU Z C, WU D H, YUE Y C.Engine wear fault diagnosis based on supervised kernel entropy component analysis[J]. Journal of system simulation, 2022, 34(1): 45-52.
[7] 彭开香, 张丽敏. 基于核典型相关性-熵成分分析的工业过程质量监测方法[J]. 控制与决策, 2021, 36(12): 2999-3006.
PENG K X, ZHANG L M.A quality monitoring method for industrial process based on kernel canonical correlation-entropy component analysis[J]. Control and decision, 2021, 36(12): 2999-3006.
[8] 孙建平, 李雪, 刘轩, 等. 改进滑动窗口核主元分析算法[J]. 热力发电, 2018, 47(1): 100-105.
SUN J P, LI X, LIU X, et al.An improved moving window kernel principal component analysis algorithm[J]. Thermal power generation, 2018, 47(1): 100-105.
[9] 郭金玉, 于欢, 李元. 基于KPCA-SVM的相关和独立变量故障检测方法[J]. 深圳大学学报(理工版), 2023, 40(1): 14-21.
GUO J Y, YU H, LI Y.Related and independent variable fault detection method based on KPCA-SVM[J]. Journal of Shenzhen University (science and engineering), 2023, 40(1): 14-21.
[10] 李榕, 申志, 李元. 基于核熵成分分析的工业过程多类型故障诊断[J]. 电子测量技术, 2023, 46(10): 40-45.
LI R, SHEN Z, LI Y.Multi-type fault diagnosis of industrial process based on KECA[J]. Electronic measurement technology, 2023, 46(10): 40-45.
[11] 张伟, 许爱强, 平殿发. 基于一种自适应核学习的KECA子空间故障特征提取[J]. 北京理工大学学报, 2017, 37(8): 863-868, 874.
ZHANG W, XU A Q, PING D F.A method for feature extraction in KECA feature subspace based on adaptive kernel learning[J]. Transactions of Beijing Institute of Technology, 2017, 37(8): 863-868, 874.
[12] 李英顺, 阚宏达, 王德彪, 等. 一种基于KPCA-WOA-SVM火控系统故障诊断方法[J]. 火炮发射与控制学报, 2023, 44(4): 14-19.
LI Y S, KAN H D, WANG D B, et al.A fault diagnosis method for fire control systems based on KPCA-WOA-SVM[J]. Journal of Gun launch & control, 2023, 44(4): 14-19.
[13] 白丽丽. 齿轮传动系统关键零部件故障状态识别方法研究[D]. 太原: 太原理工大学, 2020.
BAI L L.Research on fault condition recognition method for key parts and components of gear transmission system[D]. Taiyuan: Taiyuan University of Technology, 2020.
[14] MIRJALILI S, LEWIS A.The whale optimization algorithm[J]. Advances in engineering software, 2016, 95: 51-67.
[15] 谷晓娇, 陈长征. 基于VMD和QPSO-SR的风电机组轴承故障提取方法[J]. 太阳能学报, 2019, 40(10): 2946-2952.
GU X J, CHEN C Z.Fault extraction method of wind turbine bearing based on VMD and QPSO-SR[J]. Acta energiae solaris sinica, 2019, 40(10): 2946-2952.
[16] 徐硕, 邓艾东, 杨宏强, 等. 基于改进残差网络的旋转机械故障诊断[J]. 太阳能学报, 2023, 44(7): 409-418.
XU S, DENG A D, YANG H Q, et al.Rotating machinery fault diagnosis method based on improved residual neural network[J]. Acta energiae solaris sinica, 2023, 44(7): 409-418.
基金
河北省教育厅青年自然基金项目（QN2022185）; 邯郸学院校级项目（自然科学类XZ2019204）; 国家自然科学基金面上项目（62373151）; 国家自然科学基金联合项目（U21A20486）
PDF(1748 KB)
