<scp>PrismPatNet</scp> : Novel prism pattern network for accurate fault classification using engine sound signals

Engines are prone to various types of faults, and it is crucial detect indeed classify them accurately. However, manual fault type detection time-consuming error-prone. Automated promises reduce inter- intra-observer variability while ensuring time invariant attention during the observation duration. We have proposed an automated fault-type model based on sound signals realize these advantageous properties. named prism pattern network (PrismPatNet) reflect fact that our design incorporates a novel feature extraction algorithm was inspired by 3D shape. Our achieves high accuracy with low-computational complexity. It consists three main phases: (i) multilevel generation maximum pooling operator, (ii) ranking selection using neighbourhood component analysis (NCA), (iii) support vector machine (SVM) classification. The operator decomposes signal into six levels. extracts parameter values from both itself its decompositions. generated merged fed NCA algorithm, which 512 features input. resulting vectors passed SVM classifier, labels input as belonging 1 27 classes. validated newly collected dataset containing (1) normal engine (2) 26 different faults. reached 99.19% 98.75% 80:20 hold-out validation 10-fold cross-validation, respectively. Compared previous studies, achieved highest overall classification even though tasked identifying significantly more This performance indicates PrismPatNet ready be installed in real-world applications.