VHDL-Implemented Neuro-Fuzzy Cascade for Renewable Energy Defect Classification
International Journal of Computing · 30.06.2026 · 369-380
DOI: 10.47839/ijc.25.2.4663 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Computing, online erschienen am 30.06.2026. Verfasst von Lesia Dubchak und 1 weiteren Autor:innen.
Abstract (Originalsprache)
This paper presents a hardware-oriented neuro-fuzzy cascade for the real-time classification of wind turbine blade defects. A modified SqueezeNet model generates an informative feature vector, which is processed by Hypersector Fuzzy LVQ, Fuzzy BSB, and a modified Wang–Mendel classifier. Hypersector Fuzzy LVQ performs primary classification, whereas the switching logic evaluates the confidence margin and a hardware-efficient feature-variability indicator to select an auxiliary classifier for uncertain samples. The complete classifiers are trained offline, while their compact inference representations and switching logic are implemented in VHDL. Functional verification in Active-HDL demonstrated correct sample-dependent routing and synchronous decision generation. On the test subset, the cascade correctly classified samples, corresponding to an accuracy of 98.33%. The proposed architecture provides low decision latency after CNN feature extraction and can serve as a basis for embedded defect-monitoring systems. Its current limitations include empirical threshold selection, a relatively small test subset, and the absence of physical FPGA synthesis.
Zitieren
APA
Dubchak, L., & Rusyn, B. (2026). VHDL-Implemented Neuro-Fuzzy Cascade for Renewable Energy Defect Classification. International Journal of Computing, 369-380. https://doi.org/10.47839/ijc.25.2.4663
Vancouver
Dubchak L, Rusyn B. VHDL-Implemented Neuro-Fuzzy Cascade for Renewable Energy Defect Classification. International Journal of Computing. 2026:369-380. doi:10.47839/ijc.25.2.4663
BibTeX
@article{Dubchak2026,
author = {Dubchak, Lesia and Rusyn, Bohdan},
title = {VHDL-Implemented Neuro-Fuzzy Cascade for Renewable Energy Defect Classification},
journal = {International Journal of Computing},
year = {2026},
pages = {369--380},
doi = {10.47839/ijc.25.2.4663}
}
EUVIMED