Skip to main content

Table 5 Damage detection, case 2

From: Fully convolutional networks for structural health monitoring through multivariate time series classification

\(\varvec{f}_{\varvec{min}}\) - \(\varvec{f}_{\varvec{max}}\)(Hz) \(\varvec{\lbrace {\mathcal {F}}_{*} \rbrace }\) \(\varvec{A}^{\varvec{d}}\)
\(15-17\) \(\{{\varvec{u}}^{sh}_i\}_{i=1}^8\) 0.998
\(15-17\) \(\{{\varvec{u}}^{ax}_i\}_{i=1}^8\) 0.997
\(15-17\) \(\{{\varvec{u}}^{sh}_i\}_{i=1}^8\) and \(\{{\varvec{u}}^{ax}_i\}_{i=1}^8\) 0.999
\(5-7\) \(\{{\varvec{u}}^{sh}_i\}_{i=1}^8\) 0.996
\(5-7\) \(\{{\varvec{u}}^{ax}_i\}_{i=1}^8\) 0.892
\(5-7\) \(\{{\varvec{u}}^{sh}_i\}_{i=1}^8\) and \(\{{\varvec{u}}^{ax}_i\}_{i=1}^8\) 0.998
  1. Accuracy \({\mathbb {A}}_{d}\) of the classifier \({\mathcal {G}}_d\) evaluated on \({\mathbb {D}}^d_{test}\)