A Case-Based System For Constructing a Constitutive Equation For a Viscoelastic Material in a Laminar Flow

Abstract

The formulation of constitutive equation (model) is essential in order to simulate or predict the behavior of viscoelastic material in many complicated industrial flow processes. The generation of such model is tedious and time-consuming operation. The techniques of artificial intelligence (AI) have proven to be amenable to solving rheological problems. Several subsystems of these types have been built. They are the viscometric flow analyzer, static test analyzer and dynamic test analyzer. However, these subsystems are rather incompatible to each other and segregated. In this paper, we discuss how these subsystems can be integrated in an architecture whereby the communications between them can be controlled by a casebased system. We describe the problems commonly encountered during the experiment, the data analysis and the formulation of constitutive equation. It is then shown how the problems and solutions are presented as a case. A technique so-called model-guided repair is used for adaptation purpose in the case-based system. A schematic diagram of the architecture of the hybrid system is given in the final section.