Design of Multi-model Linear Inferential Sensors with SVM-based Switching Logic
This work addresses incremental improvements in soft sensor design for chemical engineering, focusing on enhancing model accuracy and simplicity.
The paper tackles the problem of designing multi-model linear inferential sensors by addressing discontinuities in model switching and suboptimal data labeling, resulting in a novel SVM-based training method and direct optimization approach that improves prediction accuracy in chemical engineering applications.
We study the problem of data-based design of multi-model linear inferential (soft) sensors. The multi-model linear inferential sensors promise increased prediction accuracy yet simplicity of the model structure and training. The standard approach to the multi-model inferential sensor design consists in three separate steps: 1) data labeling (establishing training subsets for individual models), 2) data classification (creating a switching logic for the models), and 3) training of individual models. There are two main issues with this concept: a) as steps 2) & 3) are separate, discontinuities can occur when switching between the models; b) as steps 1) & 3) are separate, data labelling disregards the quality of the resulting model. Our contribution aims at both the mentioned problems, where, for the problem a), we introduce a novel SVM-based model training coupled with switching logic identification and, for the problem b), we propose a direct optimization of data labelling. We illustrate the proposed methodology and its benefits on an example from the chemical engineering domain.