Characterization of Salinity Impact on Synthetic Floc Strength via Nonlinear Component Analysis

Abstract

Many complex mechanisms are inherently engaged in flocculation processes with nonlinear nature. The strength of synthetic flocculates or natural flocculates may be relevant to numerous factors as well. It will be expensive and virtually impossible to determine an exact influential list among various factors via trial and error experiments exclusively. The objective is to develop an analytical scheme for decision making about the relevant influential list at least cost. Multivariate statistical methods are actually capable of differentiating dominating factors. There is no existing research outcome being documented about applications of either principal component analysis (PCA) or nonlinear component analysis (NCA) to the whole area of flocculation and coagulation research, essentially optimization has been never achieved indeed. Compared with PCA, NCA is more versatile to solve large dimensional nonlinear multivariate problems with a potential to reach infinite dimensionality. NCA is thus proposed in a preliminary study to figure out feasibility of challenging research to extract dominating factors associated with the mechanical behavior of flocs. Without convincing evidence so far on specific utmost factor in the floc strength studies, the scale of adjustable salinity has been intentionally chosen as the first principal component to interpret variations observed in the simulation results, together with interconnections to other major principal components. Based on the pioneering methodology proposed, some interesting results are well obtained and documented. At the same time, there is no technical difficulty unquestionably to extend the proposed NCA approach to multivariable and high-dimensional nonlinear cases.