Novel Method for Identifying Gangue in Coal Mine Belt Conveyor
Efficient sorting of coal and gangue in belt conveyor is an important way to realize green mining of coal resources,and its core problem is the rapid and accurate identification of coal and gangue.Therefore,proposed a recognition method of coal and gangue in belt conveyor based on electromagnetic wave detection.Firstly,the propagation characteristics of electromagnetic signals in the detection process were analyzed by combining the reflection and transmission phenomena of electromagnetic wave signals on the surface of coal and gangue demarcation.Then,the electromagnetic signals were mapped into two-dimensional time-frequency diagrams by using continuous wavelet transform(CWT),which is used as input to train and validate the MobileNetV2 coal and gangue recognition model and output the recognition results.The experimental results show that the average accuracy and average F1 of the MobileNetV2 model can reach 97.58%and 92.75%respectively,which verifies the feasibility and validity of the proposed method for the recognition of coal and gangue in belt conveyor.
electromagnetic wave detectiondeep learningcoal and gangue recognitiontime-frequency mappingMobileNetV2
NETL
NSTL
万方数据
