一种改进的动态核主元分析故障检测方法

doi:10.11949/j.issn.0438-1157.20181411

摘要/Abstract

摘要：

针对复杂工业系统动态非线性故障检测过程精度低和计算量大的问题,提出了一种改进的动态核主元分析故障检测方法，该方法首先利用不可区分度剔除相关程度较小或者不相关变量,减少数据量，然后通过观测值扩展对筛选后的新数据构建增广矩阵，并对矩阵使用核主元分析提取变量数据的非线性空间相关特征，最后通过监测T ²和SPE 两种统计量诊断出系统发生故障及识别故障变量。仿真实验证明，该方法能对风力发电机故障进行有效监测和诊断，与KPCA方法相比，改进的动态核主元分析方法对微小故障更为敏感。

关键词: 主元分析, 核函数, 故障诊断, 动态仿真, 算法

Abstract:

To solve the problems of low accuracy and large computation in dynamic non-linear detection process of complex industrial system, a fault detection method of improved dynamic kernel principal component analysis (IDKPCA) is proposed. First, the undistinguishable degree is used to eliminate the variables with low correlation degree or no correlation degree, so as to the amount of data is reduced, then the augmented matrix is constructed for the new data after screening by extending the observed value, the nonlinear spatial correlation characteristics of variable data is extracted by KPCA, finally monitoring statistics T ² and SPE are used to diagnose system failure and identify fault variables. Simulation experiment shows that this method can effectively monitor and diagnose the fault of wind turbine, and compared with KPCA method, the improved dynamic kernel principal component analysis method is more sensitive to minor faults.

Key words: principal component analysis, kernel function, fault diagnosis, dynamic simulation, algorithm

中图分类号:

TP 206

翟坤, 杜文霞, 吕锋, 辛涛, 句希源. 一种改进的动态核主元分析故障检测方法[J]. 化工学报, 2019, 70(2): 716-722.

Kun ZHAI, Wenxia DU, Feng LYU, Tao XIN, Xiyuan JU. Fault detect method based on improved dynamic kernel principal component analysis[J]. CIESC Journal, 2019, 70(2): 716-722.

图/表 9

图1 KPCA方法

Fig.1 Schematic diagram of KPCA method

图2 径向基核函数的特性曲线

Fig.2 Characteristic curve of radial basis kernel function

图3 基于IDKPCA故障诊断流程

Fig.3 Flow chart of fault diagnosis based on IDKPCA

表1 风力发电机系统运行测量变量

Table 1 Measurement variables and corresponding numbers of wind generator system

编号	名称	编号	名称	编号	名称	编号	名称
1	桨距角1	6	电机定子V温度	11	低速轴承温度	16	风向角
2	桨距角2	7	冷却风扇进口温度	12	高速轴承温度	17	机舱位置
3	桨距角3	8	冷却风扇出口温度	13	电网频率
4	发电机转速	9	齿轮箱冷却水温度	14	机舱振动传感器X
5	有功功率	10	风轮转速	15	瞬时风速

图4 主元累计贡献率

Fig.4 Cumulative contribution rate of principal component

图5 14号变量故障的IDKPCA检测结果

Fig.5 IDKPCA detection results for variable No.14 fault

图6 14号变量故障的KPCA检测结果

Fig.6 KPCA detection results for variable No.14 fault

图7 IDKPCA和KPCA故障检测能力对比

Fig.7 Comparison of fault detection capability between IDKPCA and KPCA

图8 14号变量故障IDKPCA与KPCA贡献值比较

Fig.8 Comparison of contribution value of IDKPCA and KPCA for variable No.14 fault

参考文献 31

1	Qin S J . Process data analytics in the era of big data[J].AIChE Journal, 2014, 60(9): 3092-3100.
2	Dai X , Gao Z . From model, signal to knowledge: a data-driven perspective of fault detection and diagnosis[J]. IEEE Transactions on Industrial Informatics, 2013, 9(4): 2226-2238.
3	Wang G , Yin S . Quality-related fault detection approach based on orthogonal signal correction and modified PLS[J]. IEEE Transactions on Industrial Informatics, 2017, 11(2): 398-405.
4	Peng K , Zhang K , Li G . Online contribution rate based fault diagnosis for nonlinear industrial processes[J].Acta Automatica Sinica, 2014, 40(3): 423-430.
5	Huang H H , Yeh Y R . An iterative algorithm for robust kernel principal component analysis[J]. Neurocomputing, 2011, 74(18): 3921-3930.
6	姜万录, 吴胜强, 刘思远 . 指数加权动态核主元分析法及其在故障诊断中应用[J]. 机械工程学报, 2011, 47(3): 63-68.
	Jiang W L , Wu S Q , Liu S Y . Exponential weighted dynamic kernel pivot element analysis and its application in fault diagnosis[J]. Chinese Journal of Mechanical Engineering, 2011, 47(3): 63-68.
7	Dai C , Liu Z , Hu K , et al . Fault diagnosis approach of traction transformers in high-speed railway combining kernel principal component analysis with random forest[J]. IET Electrical Systems in Transportation, 2016, 6(3): 202-206.
8	王树东, 李军, 高翔 . 指数加权主元分析法及其在故障诊断中的应用[J]. 工业仪表与自动化装置, 2016, (6): 117-119.
	Wang S D , Li J , Gao X . Exponential weighted PCA and its application in fault diagnosis [J]. Industrial Instruments and Automation Devices, 2016, (6): 117-119.
9	吴希军, 胡春海 . 基于核主元分析与神经网络的传感器故障诊断新方法[J]. 传感技术学报, 2006, 19(1): 26-29.
	Wu X J , Hu C H . A new sensor fault diagnosis method based on kernel principal component analysis and neural network [J]. Journal of Sensing Technology, 2006, 19(1): 26-29.
10	Pozo F , Vidal Y . Wind turbine fault detection through principal component analysis and statistical hypothesis testing[J]. Advances in Science & Technology, 2016, 101(1): 45-54.
11	齐咏生, 张海利, 王林, 等 . 基于MSPCA-KECA的冷水机组故障监测及诊断[J]. 化工学报, 2017, 68(4): 1499-1508.
	Qi Y S , Zhang H L , Wang L , et al . Fault monitoring and diagnosis of chillers based on MSPCA-KECA [J]. CIESC Journal, 2017, 68(4): 1499-1508.
12	Taouali O , Jaffel I , Lahdhiri H , et al . New fault detection method based on reduced kernel principal component analysis (RKPCA)[J]. International Journal of Advanced Manufacturing Technology, 2016, 85(5-8): 1547-1552.
13	胡云鹏 . 基于主元分析的传感器故障检测盲区预测[J]. 化工学报, 2017, 68(4): 1509-1515.
	Hu Y P . Blind area prediction of sensor fault detection based on principal component analysis [J]. CIESC Journal, 2017, 68(4): 1509-1515.
14	Tipping M E , Bishop C M . Probabilistic principal component analysis[J]. Journal of the Royal Statistical Society, 2010, 61(3): 611-622.
15	Gao Y , Wang X , Wang Z , et al . Fault detection in time-varying chemical process through incremental principal component analysis[J]. Chemometrics & Intelligent Laboratory Systems, 2016, 158(15): 102-116.
16	贾明兴, 牛大鹏, 王福利, 等 . 基于RBF神经网络的非线性主元分析新方法[J]. 仪器仪表学报, 2007, 29(3): 5684-5687.
	Jia M X , Niu D P , Wang F L , et al . A new nonlinear principal component analysis method based on RBF neural network[J]. Chinese Journal of Scientific Instrument, 2007, 29(3): 5684-5687.
17	Lei C , Chen S , Yao L . Fault diagnosis and model reference tracking fault-tolerant control for the non-Gaussian nonlinear stochastic distribution control system using Takagi-Sugeno fuzzy model[C]// Proceeding of the 34th Chinese Control Conference. Hangzhou: IEEE, 2015: 1746-1751.
18	Li R , Rong G . Fault isolation by partial dynamic principal component analysis in dynamic process[J]. Chinese Journal of Chemical Engineering, 2006, 14(4): 486-493.
19	张敏龙, 王涛, 王旭平, 等 . 分步动态自回归核主元分析及其在故障诊断中应用[J]. 计算机应用, 2016, 36(5): 1464-1468.
	Zhang M L , Wang T , Wang X P , et al . Stepwise dynamic autoregressive kernel pivot element analysis and its application in fault diagnosis [J]. Computer Applications, 2016, 36(5): 1464-1468.
20	刘春燕, 于春梅 . 基于改进核主元分析的TE故障诊断[J]. 计算机测量与控制, 2016, 24(10): 36-38.
	Liu C Y , Yu C M . TE fault diagnosis based on improved kernel pivot analysis [J]. Computer Measurement and Control, 2016, 24(10): 36-38.
21	张家良, 曹建福, 高峰, 等 . 结合非线性频谱与核主元分析的复杂系统故障诊断方法[J]. 控制理论与应用, 2012, 29(12): 1558-1564.
	Zhang J L , Cao J F , Gao F , et al . Fault diagnosis method for complex systems based on nonlinear spectrum and kernel principal component analysis[J]. Control Theory and Applications, 2012, 29(12): 1558-1564.
22	张曦 . 基于核主元分析和邻近支持向量机的汽轮机凝汽器过程监控和故障诊断[J]. 中国电机工程学报, 2007, 27(14): 56-61.
	Zhang X . Monitoring and fault diagnosis of steam turbine condenser process based on kernel principal component analysis and neighboring support vector machine [J]. Proceedings of the Chinese Society for Electrical Engineering, 2007, 27(14): 56-61.
23	梁晴晴, 韩华, 崔晓钰, 等 . 基于主元分析-概率神经网络的制冷系统故障诊断[J]. 化工学报, 2016, 67(3): 1022-1031.
	Liang Q Q , Han H , Cui X Y , et al . Fault diagnosis of refrigeration system based on PCA - probabilistic neural network [J]. CIESC Journal, 2016, 67(3): 1022-1031.
24	李荣雨, 荣冈 . 基于故障映射向量和结构化残差的主元分析(PCA)故障隔离[J]. 控制理论与应用, 2008, 25(6): 1099 -1104.
	Li R Y . Rong G. Principal component analysis (PCA) of fault isolation based on fault mapping vector and structured residual[J]. Control Theory and Applications, 2008, 25( 6): 1099 - 1104.
25	Miller J P . Statistical signatures used with principal component analysis for fault detection and isolation in a continuous reactor[J]. Journal of Chemometrics, 2010, 20(1/2): 34-42
26	宋梅村, 蔡琦 . 基于动态PCA的核动力装置传感器故障检测[J]. 武汉理工大学学报(交通科学与工程版), 2012, 36(6): 1184-1187.
	Song M C , Cai Q . Sensor fault detection of nuclear power plant based on dynamic PCA [J]. Journal of Wuhan University of Technology (Traffic Science and Engineering), 2012, 36(6): 1184-1187.
27	杨沛武, 赵忠盖, 刘飞 . 动态核概率主元分析模型及其应用[J]. 清华大学学报(自然科学版), 2008, 48(s2): 130-134.
	Yang P W , Zhao Z G , Liu F . Dynamic kernel probability pivot element analysis model and its application [J]. Journal of Tsinghua University (Natural Science Edition), 2008, 48(s2): 130-134.
28	袁哲, 石怀涛 . 基于分步动态核主元分析的故障诊断方法[J]. 沈阳建筑大学学报(自然科学版), 2013, 29(6): 1092-1097.
	Yuan Z , Shi H T . Fault diagnosis method based on stepwise dynamic kernel pivot analysis [J]. Journal of Shenyang Jianzhu University (Natural Science Edition), 2013, 29(6): 1092-1097.
29	李明虎, 李钢, 钟麦英 . 动态核主元分析在无人机故障诊断中的应用[J]. 山东大学学报(工学版), 2017, 47(5): 215-222.
	Li M H , Li G , Zhong M Y . Application of dynamic kernel principal component analysis in UAV fault diagnosis [J]. Journal of Shandong University(Engineering and Science Edition), 2017, 47(5): 215-222.
30	Lee J M , Qin S J , Lee I B . Fault detection of nonlinear processes using kernel independent component analysis[J].Canadian Journal of Chemical Engineering, 2010, 85(4): 526-536.
31	邓晓刚, 田学民 . 基于免疫核主元分析的故障诊断方法[J]. 清华大学学报(自然科学版), 2008, 48(10): 100-104.
	Deng X G , Tian X M . Fault diagnosis method based on immune kernel principal component analysis[J]. Journal of Tsinghua University (Natural Science Edition), 2008, 48(10): 100-104.

[1]	杨欣, 王文, 徐凯, 马凡华. 高压氢气加注过程中温度特征仿真分析[J]. 化工学报, 2023, 74(S1): 280-286.
[2]	宋嘉豪, 王文. 斯特林发动机与高温热管耦合运行特性研究[J]. 化工学报, 2023, 74(S1): 287-294.
[3]	张思雨, 殷勇高, 贾鹏琦, 叶威. 双U型地埋管群跨季节蓄热特性研究[J]. 化工学报, 2023, 74(S1): 295-301.
[4]	曹跃, 余冲, 李智, 杨明磊. 工业数据驱动的加氢裂化装置多工况切换过渡状态检测[J]. 化工学报, 2023, 74(9): 3841-3854.
[5]	张逸豪, 王振雷. 基于最大信息系数的分组支持向量数据描述故障检测[J]. 化工学报, 2023, 74(9): 3865-3878.
[6]	尹刚, 李伊惠, 何飞, 曹文琦, 王民, 颜非亚, 向禹, 卢剑, 罗斌, 卢润廷. 基于KPCA和SVM的铝电解槽漏槽事故预警方法[J]. 化工学报, 2023, 74(8): 3419-3428.
[7]	诸程瑛, 王振雷. 基于改进深度强化学习的乙烯裂解炉操作优化[J]. 化工学报, 2023, 74(8): 3429-3437.
[8]	王光, 单发顺, 钱禹丞, 焦建芳. 基于集成学习传递熵的化工过程微小故障检测方法[J]. 化工学报, 2023, 74(7): 2967-2978.
[9]	周继鹏, 何文军, 李涛. 异形催化剂上乙烯催化氧化失活动力学反应工程计算[J]. 化工学报, 2023, 74(6): 2416-2426.
[10]	高学金, 姚玉卓, 韩华云, 齐咏生. 基于注意力动态卷积自编码器的发酵过程故障监测[J]. 化工学报, 2023, 74(6): 2503-2521.
[11]	郑书闽, 郭鹏程, 颜建国, 王帅, 李文博, 周淇. 微小通道内过冷流动沸腾阻力特性实验及预测研究[J]. 化工学报, 2023, 74(4): 1549-1560.
[12]	明迁, 高逸, 胡剑, 李盛杰, 王金江. 热交换器泄漏故障虚拟感知方法研究[J]. 化工学报, 2023, 74(4): 1836-1846.
[13]	张生安, 刘桂莲. 高效太阳能电解水制氢系统及其性能的多目标优化[J]. 化工学报, 2023, 74(3): 1260-1274.
[14]	顾学荣, 刘硕士, 杨思宇. 基于并行EGO和代理模型辅助的多参数优化方法研究[J]. 化工学报, 2023, 74(3): 1205-1215.
[15]	陈家辉, 杨鑫泽, 陈顾中, 宋震, 漆志文. 以离子液体密度为例的分子性质预测模型建模方法探讨[J]. 化工学报, 2023, 74(2): 630-641.