Prediction interval-based modelling of polymerization reactor: a new modelling strategy for chemical reactors

Precise and reliable modelling of polymerization reactor is challenging due to its complex reaction mechanism and non-linear nature. Researchers often make several assumptions when deriving theories and developing models for polymerization reactor. Therefore, traditional available models suffer from high prediction error. In contrast, data-driven modelling techniques provide a powerful framework to describe the dynamic behaviour of polymerization reactor. However, the traditional NN prediction performance is significantly dropped in the presence of polymerization process disturbances. Besides, uncertainty effects caused by disturbances present in reactor operation can be properly quantified through construction of prediction intervals (PIs) for model outputs. In this study, we propose and apply a PI-based neural network (PI-NN) model for the free radical polymerization system. This strategy avoids assumptions made in traditional modelling techniques for polymerization reactor system. Lower upper bound estimation (LUBE) method is used to develop PI-NN model for uncertainty quantification. To further improve the quality of model, a new method is proposed for aggregation of upper and lower bounds of PIs obtained from individual PI-NN models. Simulation results reveal that combined PI-NN performance is superior to those individual PI-NN models in terms of PI quality. Besides, constructed PIs are able to properly quantify effects of uncertainties in reactor operation, where these can be later used as part of the control process. © 2014 Taiwan Institute of Chemical Engineers.

Estimating hydrogen sulphide dissipation rate constant under the influence of different chemical dosing

Sewer odour and corrosion is caused by the reduction of sulphide ions and the release of hydrogen sulphide gas (H2S) into the sewer atmosphere. The reduction of sulphide is determined by its dissipation rate which depends on many processes such as emission, oxidation and precipitation that prevail in wastewater environments. Two factors that mainly affect the dissipation of sulphide are sewer hydraulics and wastewater characteristics; modification to the latter by dosing certain chemicals is known as one of the mitigation strategies to control the dissipation of sulphide. This study investigates the dissipation of sulphide in the presence of NaOH, Mg(OH)2, Ca(NO3)2 and FeCl3 and the dissipation rate is developed as a function of hydraulic parameters such as the slope of the sewer and the velocity gradient. Experiments were conducted in a 18m experimental sewer pipe with adjustable slope to which, firstly no chemical was added and secondly each of the above mentioned chemicals was supplemented in turn. A dissipation rate constant of 2×10^-6 for sulphide was obtained from experiments with no chemical addition. This value was then used to predict the sulphide concentration that was responsible for the emission of H2S gas in the presence of one of the above mentioned four chemicals. It was found that the performance of alkali substances (NaOH and Mg(OH)2) in suppressing the H2S gas emission was excellent while ferric chloride showed a moderate mitigating effect due to its slow reaction kinetics. Calcium nitrate was of little value since the wastewater used in this study experienced almost no biological growth. Thus the effectiveness of selected chemicals in suppressing H2S gas emission had the following order: NaOH ≥ Mg(OH)2 ≥ FeCl3 ≥ Ca(NO3)2.

Prediction interval-based modelling and control of nonlinear processes

 Novel computational intelligence-based methods have been investigated to quantify uncertainties prevalent in the operation of chemical plants. A new family of predication interval-based controlling algorithms is proposed and successfully applied to chemical reactors in order to minimise energy consumption and operational cost.

Peptide sequence effects control the single pot reduction, nucleation, and growth of Au nanoparticles

Peptides have demonstrated unique capabilities to fabricate inorganic nanomaterials of numerous compositions through noncovalent binding of the growing surface in solution. In this contribution, we demonstrate that these biomolecules can control all facets of Au nanoparticle fabrication, including Au3+ reduction, without the use of secondary reagents. In this regard using the AuBP1 peptide, the N-terminal tryptophan residue is responsible for driving Au3+ reduction to generate Au nanoparticles passivated by the oxidized peptide in solution, where localized residue context effects control the reducing strength of the biomolecule. The process was fully monitored by both time-resolved monitoring of the growth of the localized surface plasmon resonance and transmission electron microscopy. Nanoparticle growth occurs by a unique disaggregation of nanoparticle aggregates in solution. Computational modeling demonstrated that the oxidized residue of the peptide sequence does not impact the biomolecule's ability to bind the inorganic surface, as compared to the parent peptide, confirming that the biomolecule can be exploited for all steps in the nanoparticle fabrication process. Overall, these results expand the utility of peptides for the fabrication of inorganic nanomaterials, more strongly mimicking their use in nature via biomineralization processes. Furthermore, these capabilities enhance the simplicity of nanoparticle production and could find rapid use in the generation of complex multicomponent materials or nanoparticle assembly.