[1]任少君,肖晋飞,司风琪,等.一种抑制残差污染的补偿型自联想神经网络[J].东南大学学报(自然科学版),2020,50(4):712-720.[doi:10.3969/j.issn.1001-0505.2020.04.016]
 Ren Shaojun,Xiao Jinfei,Si Fengqi,et al.A compensation auto-associative neural network for overcoming smearing effects[J].Journal of Southeast University (Natural Science Edition),2020,50(4):712-720.[doi:10.3969/j.issn.1001-0505.2020.04.016]
点击复制

一种抑制残差污染的补偿型自联想神经网络()
分享到:

《东南大学学报(自然科学版)》[ISSN:1001-0505/CN:32-1178/N]

卷:
50
期数:
2020年第4期
页码:
712-720
栏目:
自动化
出版日期:
2020-07-20

文章信息/Info

Title:
A compensation auto-associative neural network for overcoming smearing effects
作者:
任少君肖晋飞司风琪曹越陈家乐
东南大学能源热转换及其过程测控教育部重点实验室, 南京 210096; 东南大学能源与环境学院, 南京 210096
Author(s):
Ren Shaojun Xiao Jinfei Si Fengqi Cao Yue Chen Jiale
Key Laboratory of Energy Thermal Conversion and Control of Ministry of Education, Southeast University, Nanjing 210096, China
School of Energy and Environment, Southeast University, Nanjing 210096, China
关键词:
自联想神经网络 残差污染 残差补偿 故障诊断
Keywords:
auto-associative neural network smearing effects residual compensation fault diagnosis
分类号:
TP206.3
DOI:
10.3969/j.issn.1001-0505.2020.04.016
摘要:
针对常规自联想神经网络容易陷入残差污染而难以准确重构的问题,提出了一种新的补偿型自联想神经网络算法.该算法通过在常规自联想神经网络的测试过程中引入网络补偿层,建立了网络输入层与残差空间之间的调整机制,并采用梯度下降法快速获取目标变量的残差补偿幅值.给出了单变量和多变量残差补偿量计算流程,通过比较补偿后平方预测误差(SPE)统计量的大小来确定最佳补偿方向和补偿幅值,从而定位到异常点位置,并计算出模型重构值.通过仿真算例和工程算例验证了所提算法的有效性,算例结果表明,该算法能够在未知异常点位置的情况下,有效克服大幅度异常和多点并行异常造成的残差污染影响,其诊断和重构性能明显优于常规自联想神经网络和主成分分析算法.
Abstract:
To overcome the smearing effect problem of the basic auto-associative neural network(AANN)that may reduce the reconstruction accuracy, a new compensation auto-associative neural network(CAANN)is proposed. In CAANN, a new compensation layer is introduced into the testing process of the basic AANN in order to develop an adjustment mechanism between the input layer and the residual space, and the residual compensation magnitude is calculated by the gradient descent algorithm. Then, a strategy of residual compensation for both single and multiple variable(s)is implemented to calculate the optimized compensation magnitude and direction by comparing the squared prediction error(SPE)after compensation, and further the source of anomalies is pinpointed and the reconstructed values are obtained. The effectiveness of the proposed method is evaluated on a validation example and an industrial example. The results demonstrate that the proposed CAANN can effectively inhibit the drawbacks of “smearing effects” in case of both large amplitude anomalies and multipoint concurrent anomalies without prior knowledge, showing a better performance than AANN and principal component analysis(PCA)methods for fault diagnosis and data reconstruction.

参考文献/References:

[1] Heo S, Lee J H. Parallel neural networks for improved nonlinear principal component analysis[J]. Computers & Chemical Engineering, 2019, 127: 1-10. DOI:10.1016/j.compchemeng.2019.05.011.
[2] Ghosh A, Wang G N, Lee J. A novel automata and neural network based fault diagnosis system for PLC controlled manufacturing systems[J]. Computers & Industrial Engineering, 2020, 139: 106188. DOI:10.1016/j.cie.2019.106188.
[3] Hines J W, Uhrig R E, Wrest D J. Use of autoassociative neural networks for signal validation[J]. Journal of Intelligent and Robotic Systems, 1998, 21: 143-154.
[4] Kramer M A. Autoassociative neural networks[J].Computers & Chemical Engineering, 1992, 16(4): 313-328. DOI:10.1016/0098-1354(92)80051-a.
[5] Kramer M A. Nonlinear principal component analysis using autoassociative neural networks[J].AIChE Journal, 1991, 37(2): 233-243. DOI:10.1002/aic.690370209.
[6] Gautam C, Ravi V. Counter propagation auto-associative neural network based data imputation[J]. Information Sciences, 2015, 325: 288-299. DOI:10.1016/j.ins.2015.07.016.
[7] 周建民,张臣臣,张龙,等.基于自联想神经网络与模糊C均值的滚动轴承的性能退化评估[J].机械设计与研究,2019,35(1):96-99. DOI:10.13952/j.cnki.jofmdr.2019.0107.
Zhou J M, Zhang C C, Zhang L, et al. Rolling bearing performance degradation assessment based on AANN-FCM[J]. Machine Design & Research, 2019, 35(1): 96-99. DOI:10.13952/j.cnki.jofmdr.2019.0107. (in Chinese)
[8] Xiao H J, Huang D P, Pan Y P, et al. Fault diagnosis and prognosis of wastewater processes with incomplete data by the auto-associative neural networks and ARMA model[J]. Chemometrics and Intelligent Laboratory Systems, 2017, 161: 96-107. DOI:10.1016/j.chemolab.2016.12.009.
[9] Elnour M, Meskin N, Al-Naemi M. Sensor data validation and fault diagnosis using auto-associative neural network for HVAC systems[J]. Journal of Building Engineering, 2020, 27: 100935. DOI:10.1016/j.jobe.2019.100935.
[10] Shang C, Ji H Q, Huang X L, et al. Generalized grouped contributions for hierarchical fault diagnosis with group Lasso[J]. Control Engineering Practice, 2019, 93: 104193. DOI:10.1016/j.conengprac.2019.104193.
[11] Wang J G, Cai X Z, Yao Y, et al. Statistical process fault isolation using robust nonnegative garrote[J].Journal of the Taiwan Institute of Chemical Engineers, 2020, 107: 24-34. DOI:10.1016/j.jtice.2019.12.004.
[12] Wang K, Chen J, Song Z H. A sparse loading-based contribution method for multivariate control performance diagnosis[J]. Journal of Process Control, 2020, 85: 199-213. DOI:10.1016/j.jprocont.2019.12.001.
[13] Song B, Tan S, Shi H B, et al. Fault detection and diagnosis via standardized k nearest neighbor for multimode process[J]. Journal of the Taiwan Institute of Chemical Engineers, 2020, 106: 1-8. DOI:10.1016/j.jtice.2019.09.017.
[14] Marseguerra M, Zoia A. The auto associative neural network in signal analysis: Ⅱ. Application to on-line monitoring of a simulated BWR component[J]. Annals of Nuclear Energy, 2005, 32(11): 1207-1223. DOI:10.1016/j.anucene.2005.03.005.
[15] Najafi M, Gulp C, Langari R. Enhanced auto-associative neural networks for sensor diagnostics(E-AANN)[C]//2004 IEEE International Conference on Fuzzy Systems. Budapest, Hungary, 2004, 1: 453-456. DOI:10.1109/fuzzy.2004.1375771.
[16] Abdella M, Marwala T. The use of genetic algorithms and neural networks to approximate missing data in database[C]//IEEE 3rd International Conference on Computational Cybernetics. Mauritius, 2006: 207-212. DOI:10.1109/icccyb.2005.1511574.
[17] Ravi V, Krishna M. A new online data imputation method based on general regression auto associative neural network[J].Neurocomputing, 2014, 138: 106-113. DOI:10.1016/j.neucom.2014.02.037.
[18] Sadough Vanini Z N, Meskin N, Khorasani K. Multiple-model sensor and components fault diagnosis in gas turbine engines using autoassociative neural networks[J].Journal of Engineering for Gas Turbines and Power, 2014, 136(9): 091603. DOI:10.1115/1.4027215.
[19] Palmé T, Fast M, Thern M. Gas turbine sensor validation through classification with artificial neural networks[J].Applied Energy, 2011, 88(11): 3898-3904. DOI:10.1016/j.apenergy.2011.03.047.
[20] Amirkhani S, Chaibakhsh A, Ghaffari A. Nonlinear robust fault diagnosis of power plant gas turbine using Monte Carlo-based adaptive threshold approach[J/OL].ISA Transactions, 2019 [2019-12-25]. https: //doi.org/10.1016/j.isatra.2019.11.035. DOI:10.1016/j.isatra.2019.11.035.
[21] Alobaid F, Postler R, Ströhle J, et al. Modeling and investigation start-up procedures of a combined cycle power plant[J].Applied Energy, 2008, 85(12): 1173-1189. DOI:10.1016/j.apenergy.2008.03.003.

备注/Memo

备注/Memo:
收稿日期: 2020-01-07.
作者简介: 任少君(1989—),男,博士,讲师,rsj@seu.edu.cn.
基金项目: 国家自然科学基金资助项目(51976031)、中央高校基本科研业务费专项资金资助项目(2242019K40010)、江苏高校“青蓝工程”资助项目、江苏高校优势学科建设工程资助项目.
引用本文: 任少君,肖晋飞,司风琪,等.一种抑制残差污染的补偿型自联想神经网络[J].东南大学学报(自然科学版),2020,50(4):712-720. DOI:10.3969/j.issn.1001-0505.2020.04.016.
更新日期/Last Update: 2020-07-20