跳汰机床层松散状况软测量建模方法研究

应用人工智能的方法建立了跳汰机床层松散状况的软测量模型，该软测量模型的建立分2步完成：首先根据工业现场的操作经验和化验分析结果，应用跳汰机分选效果主要指标（不完善度和错配物总量）与床层松散状况的关系建立模糊推理系统（FIS）模型，实现跳汰机床层松散状况的离线评价；然后根据浮标传感器的输出与跳汰机床层松散的关系以及跳汰机床层松散状况的离线评价结果，建立基于自适应神经模糊推理系统（ANFIS）的跳汰机床层松散状况的在线估计模型．实验得到软测量模型的训练均方根误差为0．0147，验证均方根误差为0．0214，充分体现ANFIS具有显著的学习能力和良好的泛化能力．

A Soft-sensing Model for Mobility of Jig Bed

The soft-sensing model was proposed for mobility of jig bed using the artificial intelligence,which includes two stages.Firstly,to construct the fuzzy inference system(FIS) model using the relation ship between bed mobility and separation efficiency,including imperfection and misplaced material,based on the operating experience and samples analysis,to evaluate the mobility of jig bed offline. Then,estimation model online of mobility of bed is proposed via the adaptive neuro-fuzzy inference system(ANFIS) which is built by the output of buoy sensor and the results of offline model in the first stage.The experiments show that the training and testing mean squared root error are 0.014 7 and 0.021 4,respectively,which indicates that ANFIS has remarkable abilities of learning and generalization performance.