基于多传感器信息融合双进双出磨煤机料位检测

Material Level Detection of Double-in Double-out Coal Mill Based on Multi-sensor Fusion

为解决双进双出磨煤机难于实现整个运行工况精确料位检测这一问题,提出一种基于多传感器信息融合的料位检测方法,融合系统由粗糙集和模糊神经网络来实现。根据磨煤机的工作特性将其运行工况分为三个区间,应用粗糙集理论分析不同区间中各传感器信息对于融合的重要性和决策规则的置信度,再利用粗糙集分析结果构成模糊设计网络来实现从多传感器信息到磨煤机料位的映射,并将粗糙集分析得到的属性重要度和规则置信度引入到模糊神经网络的学习过程中。通过试验结果证明研究方法的有效性,能够实现各工况较为精确的料位检测。

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.