Understanding of nonlinear environmental processes using chemometric driven model with time difference concept

Recently, there has been a growing concern over the understanding of environmental processes, since various types of phenomena, such as transport, reactions, and toxicological chemicals fate exist inside the systems and need to be managed with the stronger regulation. Therefore, to ensure the people's good health, a precise understanding on environmental processes is necessary. In this study, we proposed a new nonlinear partial least squares (NPLS) method using time difference (TD) method for understanding environmental processes. The nonlinear partial least squares using time difference (TD-NPLS) method embeds the TD method into the NPLS modeling. The NPLS can model the nonlinear inner relationships in a dimension-reduced data and the TD method captures the dynamics to correlation problem, which results in the deterioration of the predictive accuracy. The TD-NPLS method is aimed at achieving a high predictive accuracy by simultaneously capturing the characteristics of nonlinearity and dynamics in a single modeling framework. The results in the indoor air quality (IAQ) case study showed that the TD-NPLS method absolutely has a higher predictive accuracy in modeling nonlinear and dynamic processes than the LPLS, NPLS, and linear partial least squares using time difference (TD-LPLS) models. Because the proposed method can improve the understanding on the phenomena inside the processes, it resulted in more detailed modeling of the IAQ process, compared to the other methods without nonlinearity (or dynamics).