INVESTIGADORES
STEGMAYER Georgina Silvia
congresos y reuniones científicas
Título:
On-line Estimation of Industrial SBR Process Variables Using Neural Networks
Autor/es:
R. MINARI, L. GUGLIOTTA, J. VEGA, G. STEGMAYER, O. CHIOTTI
Lugar:
Buenos Aires
Reunión:
Congreso; XXII Interamerican Congress of Chemical Engineering V Argentinian Congress of Chemical Engineering; 2006
Institución organizadora:
Soc. Arg. de Química
Resumen:
This theoretical work investigates the industrial production of
Styrene-Butadiene Rubber (SBR) in a continuous reactor train operated under
steady-state condition, with the aim of developing a neural network (NN) model
for estimating several process and polymer quality variables in each reactor of
the train. The estimated variables are the monomer conversion
(x), solids content (xsol), polymer production (G), particle diameter (dp), average copolymer
composition (pS), average
molecular weights (Mn and Mw) and average branching
degree (Bn3 and Bn4). To this effect, the reaction heat is assumed
to be measured in all the reactors. The NN models are trained and validated
with a large set of simulation data generated through an available
polymerization first principles (FP) model, that was previously adjusted to
experimental results [1] and is representative of a continuous industrial plant
that involves the production of SBR grade 1502 in a train of 7 identical CSTRs,
property of Petrobras Energía S.A. (Pto. Gral. San Martín, Santa Fe, Argentina).