关键词: 粗糙集, 多传感器融合, 料位检测, 模糊神经网络, 双进双出磨煤机

Abstract: In order to solve the problem that the material level detection of double-in double-out coal mill is difficult to implement under the whole operating condition. A material level detection method based on multi-sensor fusion is proposed. The fusion system is realized by rough sets and fuzzy neural network. According to the working characteristics of double-in double-out coal mill, the operating process is divided into three stages. The rough set theory is used to analyze the importance of each sensor to fusion and the confidence of decision rule in the different stages. Then fuzzy neural network is formed on the basis of the analysis results and used to implement the mapping from the multi-sensor information to material level. Furthermore, the attribute importance and rule confidence are introduced into the learning processes of fuzzy neural network. Experiment results show that the presented method is effective and it can perform precise material level detection.

Key words: 国家青年科学基金资助项目(60905054)