Process monitoring based on neural networks is getting more and attention. Compared with classical networks, high-order have natural advantages in dealing heteroscedastic data. However, might bring the risk of overfitting, which learning both key information from original data noises or anomalies. Orthogonal constraints can greatly reduce correlations between extracted features, thereby reducin...

This paper proposes a parallel monitoring method for plant-wide processes by integrating mutual information and Bayesian inference into global-local preserving projections (GLPP)-based multi-block framework. Unlike traditional multivariate statistic process (MSPM) methods, the proposed MI-PGLPP transforms several sub-block monitoringtasks fully taking advantage of distributed First, original da...

Many multivariate statistical analysis methods and their corresponding probabilistic counterparts have been adopted to develop process monitoring models in recent decades. However, the insightful connections between them rarely studied. In this study, a generalized model (GPMM) is developed with both random sequential data. Since GPMM can be reduced various linear under specific restrictions, i...

In the process industry, an alarm system is one of important ways condition monitoring. Due to complexity and irregularity information in monitoring, there are too many false alarms current system. order solve problem designing system, this paper proposes a multivariate design method based on evidence reasoning (ER) rule, considering interval-valued reliability, which can make full use accurate...

Oscillations in the control loops indicate poor performance of loops. The occurrence oscillations process loop is quite high industry, so it needs to be reduced that can work properly. first step for oscillation reduction detection. One type difficult detect intermittent oscillation. smart factory concept encourages development detection system using machine learning by being implemented online...