Jose Munoz

Applications of maximum entropy method in capillary viscometry

Maximum entropy method (MEM) is presented as a technique for processing data obtained from capillary viscometers. The performance of MEM is assessed by comparing the viscosity versus shear rate curves generated by MEM against that obtained by the standard method based on the “Weissenberg-Rabinowitsch-Mooney equation”. In all the cases in vestigated, MEM proved to be a reliable technique in coping with the experimental noise in the capillary data.

A Performance Assessment Method for Feedback Control Loops in Chemical Processes with Distribution Outputs

Abstract Distributions of output quality variables are important in many chemical processes because they significantly affect the product quality and the process performance. A performance assessment method is proposed for feedback control loops in chemical processes with distribution outputs. The method assesses the ability of control loops to minimize the fluctuations in distribution outputs. Presently, due to long delays in obtaining measurements of distribution outputs, fluctuations in distribution outputs are controlled by a non-distribution variable. The non-distribution variable is a process variable that describes the conditions of a chemical process operation. To date, there has been no established method of estimating probability density distribution function (PDF) of a distribution from measured values of non-distribution controlled variable. In this paper, an algorithm is developed to estimate the PDF. An indicator of the performance assessment is defined in terms of the variance of the PDF of the over-all distribution output. This indicator can be used to compare various PDFs of different control loops to determine which control loop yields the optimal PDF.