Interpretation of non-linear empirical data-based process models using global sensitivity analysis

Atmospheric particulate matter characterization by Fourier Transform Infrared spectroscopy: a review of statistical calibration strategies for carbonaceous aerosol quantification in US measurement networks

“…As with complex mechanistic models, a general investigation can be carried out through sensitivity analyses (Harrington et al, 2000;Chen and Yang, 2011). One of the advantages of a PLS regression approach is that the contribution of each LV to response (y) or spectra matrix (X) can be characterized by the explained Sum-of-Squares (SS) and its normalized surrogate,…”

Model-aided optimization and analysis of multi-component catalysts: Application to selective hydrogenation of cinnamaldehyde

“…For each GPR model, global SA was implemented to calculate the SIs. It was found that 10,000 Monte Carlo samples were sufficient for accurate approximation of the integrals while being computationally manageable (Chen and Yang, 2011). The main effects due to the three components are shown in Table 3 (both S i and S T i ).…”

“…Besides the quantitative SIs, a plot of the main effects E(y|x i ) and the interactions E(y|x i , x j ) versus the process factors is useful to graphically illustrate the impact of process factors. These effects can be calculated based on the Monte Carlo method described in (Saltelli et al, 2008;Chen and Yang, 2011). An in-house Matlab program was implemented to produce the results reported in this paper.…”

“…In contrast, global SA evaluates the factors' effect within the entire design space, giving rise to a global picture of the process being investigated. Global SA has seen increasing applications in chemical engineering, in particular for the analysis of first-principles models (McRae et al, 1982;Haaker and Verheijen, 2004;Chhatre et al, 2008;Degerman et al, 2009), and to a lesser extent for data-based models (Chen and Yang, 2011).…”

Nonlinear control based on wave model of high-purity heat integrated air separation column

“…These control strategies can only work well in the neighborhood of certain operating point. However, when the operating conditions have a sharp change, most of the controller's performances become invalid [5,32,[35][36][37]. So, the tight control of HIASC not only desires an elegant control strategy but also requires an efficient control model which can accurately describe the dynamics of HIASC and the structure is simple enough for control design.…”