سال انتشار: ۱۳۹۱

محل انتشار: چهاردهمین کنگره ملی مهندسی شیمی ایران

تعداد صفحات: ۴

نویسنده(ها):

M Poursadegh – Department of Chemical Engineering, School of Chemical and Petroleum Engineering, Shiraz University, Shiraz
F Azad –
A Jahanmiri –
M.R Rahimpour, –

چکیده:

Olefin hydroformylation is one of the important intermediate processes for production of Oxoalcohol. One of the important operating problems in Oxo process is the catalyst deactivation.Industrial experience showed that the catalytic activity of rhodium/TPP complex decreases gradually in normal operating conditions, even while there are no poisons and inhibitors. In thispaper we used data from the industrial unit during the starting 822 days of the process. A generalized power law expression (GPLE) is used for predicting catalyst activity. The coefficients of GPLE have been optimized using Differential Evolution (DE) algorithm, which is a strong andeffective optimization technique. Also a model based on the adaptive neuro-fuzzy inference system (ANFIS) has been proposed for estimating catalyst activity. By comparing both models with industrial data, it was seen that ANFIS is able to predict catalyst activity more accurately than GPLE. Finally, it was observed a good agreement between two models and plant data.