高级检索

    精馏过程单变量扰动原因的智能反演

    Intelligent Inversion of Single Variable Disturbance in Distillation Processes

    • 摘要: 由于精馏过程存在大量扰动,使得扰动原因难以确定,扰动量信息诊断精度难以提高,本文对精馏过程扰动问题进行了研究。以精馏塔回流量(L)、上升蒸汽量(V)、进料组成(zF)及进料量(F)扰动为例,在单变量扰动的基础上,将物理领域的反演思想应用于精馏故障诊断,结合精馏过程内嵌机理,建立了各扰动量与欧氏距离之间的反演模型,经过运算实现扰动量的定量分析。并将此方法录入规则,以便嵌入专家系统,实现智能辨识扰动原因。本文将此方法运用到高纯度的二元精馏塔中,实现了单变量扰动原因的定量辨识,证明了该方法的有效性。

       

      Abstract: It is usually difficult to determine disturbance mechanism during distillation processes due to large number of disturbing factors, which makes it hard to improve the accuracy of fault diagnosis.In this paper, disturbance in distillation processes was studied.With reflux flow (L), vapor flow (V), feed composition (zF) and feed flow (F) as examples, an inversion method was used to establish an inversion model between the disturbance and Euclidean distances based on single variable perturbation.The disturbance could then be analyzed by mathematic calculation.This method was used as an operation rule and embedded into expert systems.The efficacy of the proposed strategy was demonstrated by a high-purity binary distillation column system.

       

    /

    返回文章
    返回