CrossRef Open Access 2022

Leak Acoustic Emission Signal Classification and Diagnosis Based on the Fractional-Order Fourier Transfer and Ensemble Learning

Peng Liu Lei Qi

Abstrak

Fractional-order Fourier transform is a representational method of the fractional Fourier domain formed by the signal rotated by any angle counterclockwise about the origin on the coordinate axis in the time-frequency plane. This paper intends to use the Fractional-order Fourier transform to process the signals collected from the acoustic emission device, and then train the results through the ensemble method of SVME, KNN and Softmax, so as to build a model that can predict size and location of leak holes in acoustic emission device. The model process has a good accuracy in predicting that whether or not the leak and the size of the leak are empty. If you only need to predict whether it leaks, the accuracy reaches 75.6 %, compared to the model trained on the original data, the classification accuracy has increased by 25.6% to 66.8%. In particular, on the Softmax classifier, the addition of FFRT increases the accuracy by more than 200%.

Penulis (2)

P

Peng Liu

L

Lei Qi

Format Sitasi

Liu, P., Qi, L. (2022). Leak Acoustic Emission Signal Classification and Diagnosis Based on the Fractional-Order Fourier Transfer and Ensemble Learning. https://doi.org/10.3233/atde220453

Akses Cepat

PDF tidak tersedia langsung

Cek di sumber asli →
Lihat di Sumber doi.org/10.3233/atde220453
Informasi Jurnal
Tahun Terbit
2022
Bahasa
en
Sumber Database
CrossRef
DOI
10.3233/atde220453
Akses
Open Access ✓