Research Article

Wear Identification of Vibration Drilling Bit Based on Improved LMD and BP Neural Network

Table 6

The influence of different network structures and neurons on the recognition results.

Network layersNumber of neurons (single signal)Recognition rateNumber of neurons (fusion of two signals)Recognition rateNumber of neurons (fusion of three signals)Recognition rate

2Force20.7917Fusion of force signal and AE signal50.833380.9166
AE30.8333Fusion of force signal and vibration signal50.7917
Vibration30.8750Fusion of AE signal and vibration signal60.9166
3Force20.8333Fusion of force signal and AE signal50.791780.9583
AE30.9166Fusion of force signal and vibration signal50.916
Vibration30.9166Fusion of AE signal and vibration signal60.9583
4Force20.8750Fusion of force signal and AE signal50.916681.0000
AE30.9583Fusion of force signal and vibration signal50.9583
Vibration30.9166Fusion of AE signal and vibration signal60.9583
5Force20.8750Fusion of force signal and AE signal50.833380.9166
AE30.8750Fusion of force signal and vibration signal50.9166
Vibration30.9166Fusion of AE signal and vibration signal60.9583