Industrial Data Denoising via Low-Rank and Sparse Representations and Its Application in Tunnel Boring Machine

The operation data of a tunnel boring machine (TBM) reflects its geological conditions and working status, which can provide critical references essential information for TBM designers operators. However, in practice, may get corrupted due to equipment failures or management errors. Moreover, the state system usually changes, results patterns that vary comparatively. This paper proposes denoising approach process data. is combined with low-rank matrix recovery (LRMR) sparse representation (SR) theory. classical LRMR model requires noise must be sparse, but sparsity cannot fully guaranteed. In proposed model, weighted nuclear norm utilized enhance components, constraint condition number applied ensure stability solution. coupled fuzzy c-means algorithm (FCM) find natural partitioning using as input. performances are illustrated through an application Shenzhen metro. Experimental show performs well denoising. different excavation status recognition accuracy improved remarkably after