基于多特征融合的煤矿带式输送机驱动系统状态评估方法研究OA

To address the challenge of comprehensively evaluating the operating state of drive systems in coal mine belt conveyors,this paper proposes a state assessment method based on multi-feature fusion.Taking the belt conveyor installed in the 31108 transport crossheading of Cuncao Tower No.2 Coal Mine as a case study,simulation samples were generated under 80%load conditions with boundaries defined by ra-ted parameters and protection thresholds.Five feature indicators were selected from four dimensions:cur-rent,vibration,temperature rise,and speed deviation.The entropy weight method is then used to construct a Health Index(HI),enabling a quantitative representation of the drive system state.The results show that the HI exhibits a pronounced monotonic decline as the drive system deteriorates,yielding well-sepa-rated grading characteristics across different health conditions.The random forest model achieves a macro-average F1 score of 95.79%,verifying the validity of the selected multi-source features and the state clas-sification results.Comparative experiments further reveal that multi-feature fusion significantly outper-forms single-feature input,and that the random forest model has good robustness under noise perturba-tion.