基于模型迁移方法的回转窑煅烧带温度软测量

Soft-Sensing for Calcining Zone Temperature in Rotary Kiln Based on Model Migration

回转窑的煅烧带温度是其控制过程中一个非常重要的参数,但煅烧带温度难以直接获取并且缺少大量的实测数据进行软测量.为了在数据较少的情况下获得准确的软测量模型,并考虑到窑头温度与煅烧温度的相似性,引入了基于过程相似性进行模型迁移的方法(PMBPS).首先采用混沌混合学习算法训练T-S模糊神经网络,对具有大量准确测量值的窑头温度建模,然后用PMBPS算法对窑头温度模型进行规划修正,获得煅烧带温度的软测量模型.仿真验证了所提出的软测量建模方法的有效性.

The calcining zone temperature in a rotary kiln is a very important process parameter to control the kiln,but the temperature is so difficult to be measured directly and no sufficient measured data are available to soft-sensing.For the purpose of obtaining the accurate soft-sensing model in case the data are less with the temperature similarity between kiln head and calcining zone taken into consideration,the process modeling based on process similarity（PMBPS） is introduced,i.e.,the rotary kiln head temperature model is developed with lots of accurately measured values involved by T-S fuzzy neural network which is trained by the chaotic hybrid learning algorithm,then the PMBPS algorithm is applied to the kiln head temperature model to correct its deviation as a model migration algorithm so as to obtain the soft-sensing model of calcining zone temperature.Simulation results verified the effectiveness of the modeling method of soft-sensing.