Fuzzy Decisions on Fluid Instabilities: Autoencoder-Based Reconstruction meets Rule-Based Anomaly Classification

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Dogga, Bharadwaj, Raju, Gibin, Louw, Wilhelm, Cohen, Kelly
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915436677824512
author Dogga, Bharadwaj
Raju, Gibin
Louw, Wilhelm
Cohen, Kelly
author_facet Dogga, Bharadwaj
Raju, Gibin
Louw, Wilhelm
Cohen, Kelly
contents Shockwave classification in shadowgraph imaging is challenging due to limited labeled data and complex flow structures. This study presents a hybrid framework that combines unsupervised autoencoder models with a fuzzy inference system to generate and interpret anomaly maps. Among the evaluated methods, the hybrid $β$-VAE autoencoder with a fuzzy rule-based system most effectively captured coherent shock features, integrating spatial context to enhance anomaly classification. The resulting approach enables interpretable, unsupervised classification of flow disruptions and lays the groundwork for real-time, physics-informed diagnostics in experimental and industrial fluid applications.
format Preprint
id arxiv_https___arxiv_org_abs_2508_05418
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Fuzzy Decisions on Fluid Instabilities: Autoencoder-Based Reconstruction meets Rule-Based Anomaly Classification
Dogga, Bharadwaj
Raju, Gibin
Louw, Wilhelm
Cohen, Kelly
Computational Engineering, Finance, and Science
Shockwave classification in shadowgraph imaging is challenging due to limited labeled data and complex flow structures. This study presents a hybrid framework that combines unsupervised autoencoder models with a fuzzy inference system to generate and interpret anomaly maps. Among the evaluated methods, the hybrid $β$-VAE autoencoder with a fuzzy rule-based system most effectively captured coherent shock features, integrating spatial context to enhance anomaly classification. The resulting approach enables interpretable, unsupervised classification of flow disruptions and lays the groundwork for real-time, physics-informed diagnostics in experimental and industrial fluid applications.
title Fuzzy Decisions on Fluid Instabilities: Autoencoder-Based Reconstruction meets Rule-Based Anomaly Classification
topic Computational Engineering, Finance, and Science
url https://arxiv.org/abs/2508.05418