罐笼偏载状态下滑动罐耳与罐道冲击模式识别

Impact Pattern Recognition between Sliding Tank Ear and Tank Track under Unbalanced Load State of the Cage

针对现有罐道故障诊断仅考虑了罐笼平衡状态,而未考虑在超深矿井中悬挂油缸失效而引起的罐笼偏载 的问题,提出一种罐笼偏载状态下滑动罐耳与罐道正常、凸起、错位3种故障的冲击模式识别方法。以罐笼横向振动信 号小波包分解后各频带能量熵、奇异值、标准差、波形指标作为原始特征集,通过邻域粗糙集约简,去除不相关和冗余 特征以获得敏感特征集,并通过布谷鸟搜索算法(CS)优化后的支持向量机模型进行模式识别。实验研究表明:该方法 比基于遗传算法(GA)、粒子群算法(PSO)、萤火虫算法(FA)优化的支持向量机分类正确率更高,达到91.7 %,对保障提 升系统偏载状态下安全运行具有着重要意义。

Current fault diagnosis of the existing tank tracks only considers the balance state of the hoist,but ignores the unbalanced load of the hoist caused by the failure of the suspended cylinders in ultra-deep mine.In this paper,a recognition method for the impact pattern between sliding tank ear and tank under the normal,bulges and dislocation conditions of the cage guide is proposed under the unbalanced load of the hoist.With the energy entropy,singular value,standard deviation and waveform index of each frequency band after the wavelet packet decomposition of the lateral vibration signal of the hoist as the original feature set,and the irrelevant and redundant features removed by neighborhood rough set,the sensitive feature set is obtained,which is used for pattern recognition based on support vector machine optimized by cuckoo algorithm(CS).The experimental study shows that compared with genetic algorithm(GA),particle swarm optimization algorithm(PSO)and firefly algorithm(FA)optimization,the support vector machine optimized by cuckoo algorithm has higher classification accuracy(91.7%)and shorter operation time,which is of great significance to ensure the safe operation of the lifting system.