Combined generative topographic mapping and graph theory unsupervised approach for nonlinear fault identification

Abstract

in Wiley Online Library (wileyonlinelibrary.com) Identifying anomalies in chemical processes is highly desirable. Usually, one relies on previous knowledge of normal and faulty samples, excluding anomalies from model training and associating deviations to faults. How reliable is such knowledge, however, is questionable, especially during atypical scenarios. Unsupervised approaches, using no labels, provide an unbiased analysis. A generative topographic mapping (GTM) and graph theory combined approach, then, is proposed for unsupervised fault identification. GTM, given its probabilistic nature, highlights system features, reducing variable dimensionality. With this information, correlation between samples is calculated. Graph theory, then, generates a network, clustering similar samples. Two anomaly cases are analyzed: an artificial dataset and Tennessee Eastman Process. Principal component analysis (PCA) and Dynamic PCA indexes Q and T 2 along GTM and graph theoryindependent monitoring methodologies are used for comparison, considering supervised and unsupervised approaches. The proposed method performed similarly to all supervised methodologies, motivating its application and developments.

Omics‐based Identification of Glycan Structures as Biomarkers for a Variety of Diseases

“…A GTM [35] is the mathematically founded counterpart of a self-organizing map. We used the GTM implementation from Escobar et al [37,38] Full Paper www.molinf.com Table 2. Points from a d-dimensional space (referred to as the data space, which is equivalent to disease signatures in our case) are projected onto a 2D map (called the latent space).…”

“…Figure 4 shows a generative topographic map (GTM) [35,36,37,38] for the 79 diseases produced from the pathway-based glycogene signatures. Collectively, these results suggest that commonalities between different diseases can be identified using pathway-based glycogene signatures.…”

“…To some extent, the map preserves the topology of the d-dimensional input data. We used the GTM implementation from Escobar et al [37,38]…”

“…Furthermore, such commonalities could form the basis for new disease markers, with structural changes in glycans as a cause of cancer metastasis being just one example. Figure 4 shows a generative topographic map (GTM) [35,36,37,38] for the 79 diseases produced from the pathway-based glycogene signatures. Some cancers, being located in the same direction as in the PCA analysis, clustered in the region with positive values on the vertical axis (GTM mode 1) in the GTM.…”

Time-varying Forgetting Factor Stochastic Gradient Algorithm for Nonlinear Systems with Color Noise

“…System identification is a powerful technique which has great potential in many applied areas such as model-based simulation, prediction and control of dynamical systems [1][2][3][4][5]. As a result, the study of system identification has attracted the attention of researchers from many scientific disciplines such as chemical processes [6], fault identification [7], biological process [8] and power system [9], etc.…”