Adaptive Soft Sensor Development for Non-Gaussian and Nonlinear Processes

Abstract

Just-in-time (JIT) adaptive soft sensors have been widely used in chemical processes because they can deal with slow-varying processes, abrupt process changes, and outliers. However, these traditional JIT algorithms including locally weighted partial least square (LW-PLS) have limitations in dealing with non-Gaussian distributed and nonlinear data. To address these issues, a modified LW-PLS-based JIT algorithm, namely, ensemble locally weighted independent component kernel partial least square (E-LW-IC-KPLS) is proposed. Its predictive performances were tested using the data generated from a numerical example and two simulated plants. Then, the results were compared to the ones resulted from LW-PLS, locally weighted kernel partial least square (LW-KPLS), and locally weighted independent component kernel partial least square (LW-IC-KPLS) algorithms. From these comparative studies, it is evident that E-LW-IC-KPLS is superior compared to its traditional counterparts concerning predictive performances. The predictive errors for E-LW-IC-KPLS are lower by 8−94% than those of LW-PLS, LW-KPLS, and LW-IC-KPLS.

Sustainable fashion: Design of the experiment assisted machine learning for the environmental-friendly resin finishing of cotton fabric

“…With that in mind, a smaller RMSE indicates higher accuracy and prediction ability. Equation (1) displays the RMSE formula [ 22 ]. …”

A data‐driven soft sensor based on weighted probabilistic slow feature analysis for nonlinear dynamic chemical processes

“…Advanced science and technology not only bring great convenience for complex and large‐scale enterprise production but also put forward higher requirements for process control stability, detection timeliness, and operation reliability. In order to monitor the running state of the system in time and realize the smooth control of the process, the accurate measurement of product quality variables is very important 1–4 . In practice, it is very difficult to obtain the key quality indexes in time due to the bad measuring environment, expensive measuring instruments, and measurement lag.…”

A Novel Dynamic Just-in-Time Learning Framework for Modeling of Batch Processes

“…Though these techniques can handle nonlinearity, these modeling approaches assume that operating conditions during the process remain unchanged. There have also been attempts in the literature to employ the above-discussed methods into the JIT framework to address the issues related to time-varying dynamics. − For example, PLS when integrated into the JIT framework results in locally weighted partial least-squares (LWPLS). , Zhang et al have proposed a locally weighted kernel PLS (LW-KPLS) based on sparse nonlinear features which outperformed the LW-PLS by efficiently dealing with strong nonlinearity. Chen et al have introduced recursive locally weighted partial least-squares (RLWPLS) into the JIT framework to handle both the time-varying and nonlinearity issues in soft-sensor development.…”