Research on Coal Gangue Recognition Based on Multi-source Time–Frequency Domain Feature Fusion

The over-exploitation of resources caused by the increasing coal demand has resulted in a sharp increase solid waste emissions mainly gangue, which made burden on environment, economy, resources, and society our country heavier. In order to achieve balance between energy consumption emission process top caving, this study carried out gangue recognition research based multi-source time–frequency domain feature fusion (MS-TFDF-F). First, symbiosis harm caving are analyzed, fundamental method comprehensive treatment is put forward, accurate interface. Second, building simulation test bed, MS signals generated mixture with content 0–100% collected TFDFs extracted. Third, MS-TFDF-F-based model established. Then, effect two TFDF-F sample sets was compared, results show that selection (TFDFS-FM) higher accuracy. On basis, paper studies variation law number sensors accuracy information fusion. Finally, economic, social, environmental, resource benefits qualitatively described. final strongest ability when fusing six sensor signals, reaches 99% under AdaBoost algorithm. establishment brings huge China’s society, it helpful realize loss reduction mining caving.