Prediction interval-based neural network modelling of polystyrene polymerization reactor - a new perspective of data-based modelling

In this paper, prediction interval (PI)-based modelling techniques are introduced and applied to capture the nonlinear dynamics of a polystyrene batch reactor system. Traditional NN models are developed using experimental datasets with and without disturbances. Simulation results indicate that traditional NNs cannot properly handle disturbances in reactor data and demonstrate a poor forecasting performance, with an average MAPE of 22% in the presence of disturbances. The lower upper bound estimation (LUBE) method is applied for the construction of PIs to quantify uncertainties associated with forecasts. The simulated annealing optimization technique is employed to adjust NN parameters for minimization of an innovative PI-based cost function. The simulation results reveal that the LUBE method generates quality PIs without requiring prohibitive computations. As both calibration and sharpness of PIs are practically and theoretically satisfactory, the constructed PIs can be used as part of the decision-making and control process of polymerization reactors. © 2014 The Institution of Chemical Engineers.

The addition of fluoxetine to cognitive behavioural therapy for youth depression (YoDA-C): study protocol for a randomised control trial.

The aim of the Youth Depression Alleviation-Combined Treatment (YoDA-C) study is to determine whether antidepressant medication should be started as a first-line treatment for youth depression delivered concurrently with psychotherapy. Doubts about the use of medication have been raised by meta-analyses in which the efficacy and safety of antidepressants in young people have been questioned, and subsequent treatment guidelines for youth depression have provided only qualified support.

The value of visual fleece assessment in addition to objective measurements in identifying Angora goats of greater clean mohair production

Visual assessment of the fleece of Merino sheep is an accepted method to aid genetic improvement but there is little evidence to support the use of visual assessment for improving mohair production. This paper examines the extent that visual traits, including staple length, character (staple crimp), staple definition, tippiness, style and staple entanglement, are related to clean fleece weight in animals of similar live weight and mean fibre diameter (MFD) from the same flock. Measurements were made over 9 shearing periods on a population of castrated Angora males (wethers) goats representing the current range and diversity of genetic origins in Australia, including South African, Texan and interbred admixtures of these and Australian sources (these different genetic origins are defined as Breed in this work). Data on genetic origin, sire, dam, lifetime characteristics (date of birth, dam age, birth weight, birth parity (single or twin), weaning weight), live weight, fleece growth and visual fleece attributes were recorded. A restricted maximum likelihood (REML) model was developed to relate clean fleece weight with age, MFD, average fleece-free live weight, lifetime characteristics and visual fleece attributes. There were separate linear responses of clean fleece weight to MFD and staple length for each age group, a quadratic response to the square root of average fleece-free live weight, an effect of sire breed and linear responses to dam age, staple definition score and character. Depending on age at shearing, the increase in clean fleece weight was between about 50 and 80. g for each increase of 1. μm in MFD. At similar MFD, clean fleece weight was generally greater at summer shearings compared with winter shearings. There was a strong increase in clean fleece weight with average fleece-free live weight up to around 50. kg but little response in clean fleece weight for animals larger than 50. kg. There was some evidence of a smaller increase in clean fleece weight as the age of dam increased. There was an effect of Breed in the model but this effect disappeared when a random sire effect was included in the model. There was a positive response to staple length at some age groups but the response did not differ from zero in other age groups. This response varied from negligible to about 70. g per 1. cm increase in staple length. Clean fleece weight increased about 40. g per unit increase in staple definition score and increased about 30. g for every 4 units increase in the number of staple crimps. There was no evidence that clean fleece weight was affected by staple style, staple tip score or staple entanglement score or lifetime factors such as birth weight, date of birth, birth parity, or weaning weight. The results show that using a combination of measuring MFD and visually assessing the fleece for staple length, staple definition and crimps can help identify the most profitable Angora goats. In this process, the objective measurement of MFD appears essential. Visual assessment will provide some extra benefit in identifying these animals above that provided by measuring MFD alone. Animal size should be considered by mohair producers when identifying more productive mohair producing animals. © 2014 Elsevier B.V.

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.

Performance analysis of three advanced controllers for polymerization batch reactor: an experimental investigation

The performances of three advanced non-linear controllers are analyzed for the optimal set point tracking of styrene free radical polymerization (FRP) in batch reactors. The three controllers are the artificial neural network-based MPC (NN-MPC), the artificial fuzzy logic controller (FLC) as well as the generic model controller (GMC). A recently developed hybrid model (Hosen et al., 2011a. Asia-Pac. J. Chem. Eng. 6(2), 274) is utilized in the control study to design and tune the proposed controllers. The optimal minimum temperature profiles are determined using the Hamiltonian maximum principle. Different types of disturbances are introduced and applied to examine the stability of controller performance. The experimental studies revealed that the performance of the NN-MPC is superior to that of FLC and GMC. © 2013 The Institution of Chemical Engineers.

RAFT controlled synthesis of graphene/polymer hydrogel with enhanced mechanical property for pH-controlled drug release

A pH-sensitive, mechanically strong and thermally stable graphene/poly (acrylic acid) (graphene/PAA) hydrogel was prepared via reversible addition fragmentation transfer (RAFT) polymerizations in the presence of a cross-linking agent. The RAFT agent was covalently coupled onto graphene basal planes via an esterification reaction, with benzoic acid functionalities pre-attached on graphene with its aryl diazonium salt precursor. AFM and SEM analysis revealed the successful preparation of single layered graphene sheets and graphene/polymer hydrogels with pH controlled porous structures. Attenuated total reflection infrared (ATR-IR) and thermogravimetric analyzer (TGA) verified the successful stepwise preparation of graphene/PAA hydrogel. This graphene/PAA hydrogel was pH-sensitive and more mechanically elastic than the PAA hydrogel prepared without graphene. The pH sensitivity of the hydrogel was further utilized for controlled drug release. Doxorubicin was chosen as a model drug and loaded into the hydrogels. The drug loading and release experiment indicated that this hydrogel can be used to efficiently control drug release in the intestine environment (pH = 7.4), better than release in a more acidic environment.© 2013 Elsevier Ltd. All rights reserved.

Wettability investigation of UV/oand acid functionalized MWCNT and MWCNT/PMMA nanocomposites by contact angle measurement

The dispersion state of individual MWCNT in the polymer matrix influences the mechanical, thermal, and electrical properties of the resulting composite. One method of obtaining a good dispersion state of MWCNT in a polymer matrix is to functionalize the surface of MWCNT using various treatments to enhance the surface energy and increase the dispersibility of MWCNT. In this study, wettability and surface energy of UV/Oand acid-treated multiwall carbon nanotubes (MWCNTs) and its polymethyl methacrylate (PMMA) polymer nanocomposites were measured using contact angle analysis in various solvent media. Contact angle analysis was based on ethylene glycol-water-glycerol probe liquid set and data was further fitted into geometric mean (Fowkes), van Oss-Chaudhury-Good (GvOC), and Chang-Qing-Chen (CQC) models to determine both nonpolar and acid base surface energy components. Analysis was conducted on MWCNT thin films subjected to different levels of UV/Oand acid treatments as well as their resulting MWCNT/PMMA nanocomposites. Contact angle analysis of thin films and nanocomposites revealed that the total surface energy of all samples was well fitted with each other. In addition, CQC model was able to determine the surface nature and polarity of MWCNT and its nanocomposites. Results indicated that the wettability changes in the thin film and its nanocomposites are due to the change in surface chemistry. Finally, electrical properties of nanocomposites were measured to investigate the effect of surface functionality (acid or basic) on the MWCNT surfaces.

Ionic conductivity and molecular dynamic behavior in supramolecular ionic networks; the effect of lithium salt addition

Supramolecular ionic networks combine singular properties such as self-healing behaviour and ionic conductivity. In this work we present an insight into the ionic conductivity and molecular dynamic behaviour of an amorphous and semicrystalline supramolecular ionic networks (iNets) that were synthesised by self-assembly of difunctional imidazolium dicationic molecules coupled with (trifluoromethane-sulfonyl) imide dianionic molecules. Relatively low ionic conductivity values were obtained for the semicrystalline iNet below its melting point (Tm =101°C) in comparison with the amorphous iNet for which the conductivity significantly increased (~3 orders of magnitude) above 100°C. Upon LiTFSI doping, the semicrystalline iNet reached conductivity values ~ 10-3 Scm-1 due to enhanced mobility of the network which was supported by solid-state static NMR. Furthermore, the overlapping of 19F and 7Li resonance lines from both the semicrystalline network and the LiTFSI suggests fast molecular motions.

Characterization of the phase behaviour of a novel polymerizable lyotropic ionic liquid crystal

The development of new polymerizable lyotropic liquid crystals (LLCs) utilizing charged amphiphilic molecules such as those based on long chain imidazolium compounds, is a relatively new design direction for producing robust membranes with controllable nano-structures. Here we have developed a novel polymerizable ionic liquid based LLC, 1-hexadecyl-3-methylimidazolium acrylate (C16mimAcr), where the acrylate anion acts as the polymerizable moiety. The phase behaviour of the C16mimAcr upon the addition of water was characterized using small and wide angle X-ray scatterings, differential scanning calorimetry and polarized optical microscopy. We compare the phase behaviour of this new polymerizable LLC to that of the well known LLC chloride analogue, 1-hexadecyl-3-methylimidazolium chloride (C16mimCl). We find that the C16mimAcr system has a more complex phase behaviour compared to the C16mimCl system. Additional lyotropic liquid crystalline mesophases such as hexagonal phase (H1) and discontinuous cubic phase (I1) are observed at 20 °C for the acrylate system at 50 and 65 wt% water respectively. The appearance of the hexagonal phase (H1) and discontinuous cubic phase (I1) for the acrylate system is likely due to the strong hydrating nature of the acrylate anion, which increases the head group area. The formation of these additional mesophases seen for the acrylate system, especially the hexagonal phase (H1), coupled with the polymerization functionality offers great potential in the design of advanced membrane materials with selective and anisotropic transport properties.

Graphene/tri-block copolymer composites prepared via RAFT polymerizations for dual controlled drug delivery via pH stimulation and biodegradation

A novel tri-block copolymer poly(oxopentanoate ethyl methacrylate)-block-poly(pyridyl disulfide ethyl acrylate)-block-poly(ethylene glycol acrylate) [poly(OEMA-b-PDEA-b-PEGA)], retaining active keto groups and pyridyl disulfide (PDS) side functionalities, was synthesized as a drug delivery vehicle using reversible addition-fragmentation chain transfer (RAFT) polymerization method. One mimic drug pyridine-2-thione (PT) was introduced into the monomer, PDEA for copolymerization. The other mimic drug O-benzylhydroxylamine (BHA) was conjugated with tri-block copolymer via efficient oxime coupling chemistry, followed by the attachment onto graphene via π-π stacking interaction to obtain a graphene/tri-block copolymer composite. 1H NMR, UV-vis absorption spectroscopy, fluorescence spectroscopy, gel permeation chromatography (GPC), atomic force microscope (AFM) and transmission electron microscope (TEM) were used to verify the successful step-wise preparation of the tri-block copolymer and drug loaded composite. In vitro release behaviors of BHA and PT from graphene/tri-block copolymer composite via dual drug release mechanisms were investigated. BHA can be released under acid environment, while PT will be released in the presence of reducing agents, such as dithiothreitol (DTT) or glutathione (GSH). It can be envisioned that this novel composite could be exploited as a novel intracellular drug delivery system via dual release mechanisms.

Properties of HIV-1 associated cholesterol in addition to raft formation are important for virus infection

The overall HIV-1 membrane lipid contents resemble lipid rafts, and we have previously demonstrated that raft-promoting properties of virus-associated cholesterol (with modifications in either the 3β-OH group or AB rings) are important for HIV-1 infectivity. As cholesterol is present in both rafts and non-rafts domains of HIV-1 membrane, we question whether the interpretation of rafts property of virus-associated cholesterol being an absolute requirement for HIV-1 function is too simplistic. The carbon side chain of cholesterol is the third component of cholesterol that can affect the fluidity of membrane depending on its context within the lipid membrane bilayers. In this work, we have used synthetic cholesterol analogues that have different lengths of carbon side chain for our investigation. In contrast to our previous report, we have found that cholesterol side chain analogues that lack in vitro defined raft promoting-property is able to support HIV-1 replication. More specifically, cholesterol analogues with side chains of intermediate length have greater capacity to support HIV-1 infection, suggesting HIV-1 is able to maintain function using cholesterol variants that promote a range of non-rafts- to rafts-properties. Our data demonstrate cholesterol properties other than raft-promoting function also contribute to the infectivity of HIV-1.

The addition of depression to the Framingham Risk Equation model for predicting coronary heart disease risk in women

BACKGROUND: Depression is widely considered to be an independent and robust predictor of Coronary Heart Disease (CHD), however is seldom considered in the context of formal risk assessment. We assessed whether the addition of depression to the Framingham Risk Equation (FRE) improved accuracy for predicting 10-year CHD in a sample of women.

DESIGN: A prospective, longitudinal design comprising an age-stratified, population-based sample of Australian women collected between 1993 and 2011 (n=862).

METHODS: Clinical depressive disorder was assessed using the Structured Clinical Interview for Diagnostic and Statistical Manual of Mental Disorders (SCID-I/NP), using retrospective age-of-onset data. A composite measure of CHD included non-fatal myocardial infarction, unstable angina coronary intervention or cardiac death. Cox proportional-hazards regression models were conducted and overall accuracy assessed using area under receiver operating characteristic (ROC) curve analysis.

RESULTS: ROC curve analyses revealed that the addition of baseline depression status to the FRE model improved its overall accuracy (AUC:0.77, Specificity:0.70, Sensitivity:0.75) when compared to the original FRE model (AUC:0.75, Specificity:0.73, Sensitivity:0.67). However, when calibrated against the original model, the predicted number of events generated by the augmented version marginally over-estimated the true number observed.

CONCLUSIONS: The addition of a depression variable to the FRE equation improves the overall accuracy of the model for predicting 10-year CHD events in women, however may over-estimate the number of events that actually occur. This model now requires validation in larger samples as it could form a new CHD risk equation for women.