摘要
针对采用振动法对球磨机料位测量时,其特征值存在非线性和随机性强的问题,引入二型模糊逻辑对球磨机料位进行概念表示,建立了基于区间二型T-S模糊系统的球磨机料位预测模型。首先,用模糊C均值算法对数据空间进行划分,提取前件参数,然后用最小二乘法辨识后件参数,并通过反向传播算法调整前件参数,最后利用区间二型T-S模糊系统推理实现球磨机料位的软测量。实验结果表明,基于区间二型模糊系统建立的预测模型具有较高的测量精度并能更好地跟踪实际料位曲线,性能优于其他几种常用方法。
According to the existence of the strongly nolinear and random characteristics in the process of measuring fill level of ball mill by analyzing vibration signals,the type-2 fuzzy logic was introduced to represent the concepts of fill level in ball mill.An interval type-2 T- S fuzzy logic model was built to predict the fill level of the ball mill. Firstly,a fuzzy C-means clustering algorithm was used to partition the space of the data and compute the parameters of the antecedents. Then a least square method was used to identify the parameters of the consequents. Simultaneously,the back-propagation algorithm was used for tuning parameters of the antecedents. At last,the soft sensor of the fill level was realized by uncertainty reasoning based on interval Type-2 T- S fuzzy logic system. The experiment results show that the forecasting model based on it has better performance than other methods because it has higher prediction accuracy and better tracking characteristics for real fill level curve.
出处
《仪表技术与传感器》
CSCD
北大核心
2015年第12期103-106,109,共5页
Instrument Technique and Sensor
基金
国家自然科学基金资助项目(61450011)
山西省自然科学基金项目(2015011052)
关键词
球磨机料位
区间二型模糊系统
模糊C均值
最小二乘法
反向传播算法
ball mill fill level
interval type-2 fuzzy system
fuzzy C-means
least square method
back-propagation algorithm
