Development of Soft Sensor Methods Based on Wavelength Region Selection Methods

Abstract

Soft sensors have been widely used in industrial plants to estimate process variables that are difficult to measure online (Figure 1). Soft sensor models predicting an objective variable should be constructed with only important explanatory variables in terms of predictive ability, better interpretation of models and lower measurement costs. Besides, some process variables can affect an objective variable with time-delays. We therefore have proposed the methods for selecting important process variables and optimal time-delays of each variable simultaneously, by modifying the wavelength selection methods (Figure 3, 4) in spectrum analysis. The proposed methods can select time-regions of process variables as a unit by using process data that includes process variables that are delayed for a duration ranging from 0 through some decided time. A case study with real industrial data confirmed that predictive, easy-to-interpret, and appropriate models were constructed using the proposed methods (Table 2, 3, Figure 11, 12).