基于CEEMDAN和多尺度排列熵的球磨机负荷识别方法

Load Identification Method for Ball Mills based on CEEMDAN and Multi-scale Permutation Entropy

针对球磨机负荷特征提取难以及负荷状态识别难的问题,将多尺度排列熵引入到球磨机负荷识别中,提出一种具有自适应噪声的完整集成经验模态分解(CEEMDAN)与多尺度排列熵(MPE)相结合的球磨机负荷识别方法。首先,采用CEEMDAN方法对球磨机振动信号进行处理,将其分解成一系列的模态分量,选取与原始信号相关性较高的敏感模态分量进行信号重构;其次,确定多尺度排列熵算法的最优参数,根据算法得到重构信号的排列熵值;最后,计算多尺度排列熵的偏均值,以偏均值作为特征对球磨机负荷状态进行识别。实验结果表明,该方法能够有效地识别出球磨机的不同负荷状态,并具有一定的可行性。

It is difficult to extract the load characteristics and identify the loading condition for ball mills. To solve this problem, a ball mill load identification method based on Complete Ensemble Empirical Mode Decomposition with Adaptive Noise（CEEMDAN）and multi-scale permutation entropy is proposed. Firstly, the CEEMDAN is used to process the vibration signal of the ball mill and decompose it into a series of modal components, and the sensitive modal components with high correlation with the original signal are selected for signal reconstruction. Secondly, the optimal parameters for multi-scale permutation entropy algorithm are determined. Based on the algorithm, the permutation entropy of the reconstructed signal is obtained. Finally, the partial mean value of the multi-scale permutation entropy is calculated and the load state of the ball mill is identified with the partial mean value as the characteristic value. The experimental results show that this method can effectively identify the different load state of the ball mills and has a certain feasibility.