Figure 6.71. AHPCA model (with three PCs) statistics, (a) biplots of 37 reference runs, (b) cumulative explained variance with one, two and three PCs.

an AHPCA model as shown in Eqs. 6.126-6.128 after proper equalization/ synchronization, unfolding and scaling, resulting in multivariate SPM charts (Figure 6.72). Both SPE and T2 charts signal on-line when the process is out-of-control. The variable responsible for this deviation from NO is diagnosed to be variable 2 (agitator power input) by using on-line contribution plots to SPE and T2 in Figures 6.72(b) and 6.72(d). Contributions are calculated for the interval of the fault occurrence (140th and 180th measurements). Variables 7, 13 and 14 (biomass concentration, heat generated and cooling water flow rate) are also contributing to deviation. This is due to the decrease in oxygen transfer rate during that short interval. □



350 r

300 r LU 1 Q-250k CO

Was this article helpful?

0 0

Post a comment