English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Timely detected anomalies in the chemical technological processes, as well asthe earliest detection of the cause of the fault, significantly reduce theproduction cost in the industrial factories. Data on the state of thetechnological process and the operation of production equipment are received bya large number of different sensors. To better predict the behavior of theprocess and equipment, it is necessary not only to consider the behavior of thesignals in each sensor separately, but also to take into account theircorrelation and hidden relationships with each other. Graph-based datarepresentation helps with this. The graph nodes can be represented as data fromthe different sensors, and the edges can display the influence of these data oneach other. In this work, the possibility of applying graph neural networks tothe problem of fault diagnosis in a chemical process is studied. It wasproposed to construct a graph during the training of graph neural network. Thisallows to train models on data where the dependencies between the sensors arenot known in advance. In this work, several methods for obtaining adjacencymatrices were considered, as well as their quality was studied. It has alsobeen proposed to use multiple adjacency matrices in one model. We showedstate-of-the-art performance on the fault diagnosis task with the TennesseeEastman Process dataset. The proposed graph neural networks outperformed theresults of recurrent neural networks.

Graph Neural Networks with Trainable Adjacency Matrices for Fault Diagnosis on Multivariate Sensor Data