Details

Autor(en) / Beteiligte

• Si, Lei
• Wang, Zhongbin
• Liu, Xinhua
• Tan, Chao

Titel

A sensing identification method for shearer cutting state based on modified multi-scale fuzzy entropy and support vector machine

Ist Teil von

• Engineering applications of artificial intelligence, 2019-02, Vol.78, p.86-101

Ort / Verlag

Elsevier Ltd

Erscheinungsjahr

2019

Link zum Volltext

Quelle

Elsevier ScienceDirect Journals Complete

Beschreibungen/Notizen

• Accurate identification of shearer cutting state is a prerequisite for achieving safe and efficient production in coal mines. In this paper, a novel state diagnosis method is put forward based on modified multi-scale fuzzy entropy (MMFE) and support vector machine (SVM). On the basis of multi-scale entropy and fuzzy entropy, MMFE is designed to obtain stable and accurate estimation for short time series over a range of scales. Therefore, MMFE is employed to extract the feature information of vibration signals of shearer rocker arm and the complexity of time series can be reasonably embodied through some simulation analysis. Besides, the Fisher score (FS) method is utilized to sort the obtained features according to their importance and the first five features with the most important information are selected as the feature vectors. Subsequently, an improved fruit fly optimization algorithm (IFOA) is presented to optimize the parameters of SVM and the IFOA–SVM based multi-classifier is constructed to fulfill an automatic state identification. The experiment results indicate that the proposed state identification method is outperforming others and can effectively distinguish different cutting states of shearer with different operation conditions. •A novel state identification method based on MMFE, FS, IFOA and SVM is proposed.•MMFE is proposed to avert the drawbacks existing in MSE and MFE.•IFOA is put forward to achieve the parameters optimization of SVM.•Simulation and experimental analysis proved the effectivity and superiority.