Fault variables recognition using improved k-nearest neighbor reconstruction

Fault detection method using k nearest neighbor rule has shown its advantages in dealing with nonlinear, multi-mode, and nonGaussian distributed data. Once a fault is detected in industrial processes, recognizing fault variables becomes the crucial task subsequently. Recently, the method of fault variables recognition using k nearest neighbor reconstruction (FVR-kNN) has been reported. However, the way of variables estimation in FVR-kNN is not accurate when multiple variables were affected by the fault. To overcome this problem, an improved FVR-kNN method is presented in this paper. The proposed method improves the estimation step of FVR-kNN when predicting multiple variables. It can guarantee that the estimations of these potential fault variables have no mutual influence. Therefore, the proposed method can give more accurate reconstruction samples which will increase the reliability of fault variables recognition. A numerical example illustrates the effectiveness of the proposed method.

[1]  Steven X. Ding,et al.  A Review on Basic Data-Driven Approaches for Industrial Process Monitoring , 2014, IEEE Transactions on Industrial Electronics.

[2]  Si-Zhao Joe Qin,et al.  Survey on data-driven industrial process monitoring and diagnosis , 2012, Annu. Rev. Control..

[3]  P. Miller,et al.  Contribution plots: a missing link in multivariate quality control , 1998 .

[4]  Jianchang Liu,et al.  Fault diagnosis using kNN reconstruction on MRI variables , 2015 .

[5]  Zhiqiang Ge,et al.  Data Mining and Analytics in the Process Industry: The Role of Machine Learning , 2017, IEEE Access.

[6]  Jin Wang,et al.  Fault Detection Using the k-Nearest Neighbor Rule for Semiconductor Manufacturing Processes , 2007, IEEE Transactions on Semiconductor Manufacturing.

[7]  Yuan Li,et al.  Diffusion maps based k-nearest-neighbor rule technique for semiconductor manufacturing process fault detection , 2014 .

[8]  Chenglin Wen,et al.  Fault Isolation Based on k-Nearest Neighbor Rule for Industrial Processes , 2016, IEEE Transactions on Industrial Electronics.

[9]  Chenglin Wen,et al.  Fault Detection Using Random Projections and k-Nearest Neighbor Rule for Semiconductor Manufacturing Processes , 2015, IEEE Transactions on Semiconductor Manufacturing.

[10]  G Verdier,et al.  Adaptive Mahalanobis Distance and $k$ -Nearest Neighbor Rule for Fault Detection in Semiconductor Manufacturing , 2011, IEEE Transactions on Semiconductor Manufacturing.

[11]  Furong Gao,et al.  Review of Recent Research on Data-Based Process Monitoring , 2013 .

[12]  Jin Wang,et al.  Large-Scale Semiconductor Process Fault Detection Using a Fast Pattern Recognition-Based Method , 2010, IEEE Transactions on Semiconductor Manufacturing.