A5022879914- Fault Detection and Control Systems
- Spectroscopy and Chemometric Analyses
- Advanced Control Systems Optimization
- Mineral Processing and Grinding
- Advanced Statistical Process Monitoring
- Crystallization and Solubility Studies
- Particle physics theoretical and experimental studies
- Water Quality Monitoring and Analysis
- Control Systems and Identification
- Spectroscopy Techniques in Biomedical and Chemical Research
- Quantum Chromodynamics and Particle Interactions
- Iterative Learning Control Systems
- Neural Networks and Applications
- Neutrino Physics Research
- Target Tracking and Data Fusion in Sensor Networks
- Analytical Chemistry and Chromatography
- Local Government Finance and Decentralization
- Process Optimization and Integration
- Family and Disability Support Research
- Chromatography in Natural Products
- Down syndrome and intellectual disability research
- Schizophrenia research and treatment
- High-Energy Particle Collisions Research
- Autism Spectrum Disorder Research
- Child and Adolescent Health
Newcastle University
2008-2021
Center for International Environmental Law
2018-2019
University of Strathclyde
2008-2019
University College London
2017
University of Kent
2017
The Ohio State University
2007-2011
International Union of Pure and Applied Chemistry
2011
University of Buckingham
2010
University of Leeds
2008
Mines Saint-Étienne
2008
This paper presents a novel nonlinear hybrid modeling approach aimed at obtaining improvements in model performance and robustness to new data the optimal control of batch MMA polymerization reactor. The contains simplified mechanistic that does not consider gel effect stacked recurrent neural networks. Stacked networks are built characterize effect, which is one most difficult parts modeling. Sparsely sampled on polymer quality were interpolated using cubic spline function generate for...
When analyzing complex mixtures that exhibit sample-to-sample variability using spectroscopic instrumentation, the variation in optical path length, resulting from physical variations inherent within individual samples, will result significant multiplicative light scattering perturbations. Although a number of algorithms have been proposed to address effect scattering, each has associated with it underlying assumptions, which necessitates additional information relating spectra being...
Summary The primary goal of multivariate statistical process performance monitoring is to identify deviations from normal operation within a manufacturing process. basis the schemes historical data that have been collected when running under operating conditions. These are then used establish confidence bounds detect onset deviations. In contrast with traditional approaches based on Gaussian assumption, this paper proposes application infinite mixture model (GMM) for calculation bounds,...
The in situ measurement of solution supersaturation associated with the batch cooling crystallization l-glutamic acid (LGA) at 500 mL and 20 L scale sizes is assessed via ATR-FTIR spectroscopy. A partial least squares chemometric calibration model was developed for online prediction LGA concentration from measured FTIR absorbance spectra overcoming some significant challenges related to low sensitivity mid-IR frequency range, its solubility water, complex speciation chemistry. data water...
The development of reliable multivariate calibration models for spectroscopic instruments in on-line/in-line monitoring chemical and bio-chemical processes is generally difficult, time-consuming costly. Therefore, it preferable if can be used an extended period, without the need to replace them. However, many process applications, changes instrumental response (e.g. owing a change spectrometer) or variations measurement conditions temperature) cause model become invalid. In this...
A technique for fault localization in batch processes using progressive principal component analysis (PCA) modeling is proposed this paper. PCA model developed from normal process operation data and used online monitoring. Once a detected by the model, variables that are related to identified contribution analysis. The time information on when abnormalities occurred these series plot of squared prediction errors (SPE) variables. These then removed another remaining If faulty cannot be new...
There is an increasing interest in using Raman spectroscopy to identify polymorphic forms and monitor phase changes pharmaceutical products for quality control. Compared with other analytical techniques the identification of polymorphs such as X-ray powder diffractometry infrared spectroscopy, FT-Raman has advantages enabling fast, situ, nondestructive measurements complex systems suspension samples. However, samples, intensities depend on analyte concentrations well particle size, overall...
Online powder X-ray diffraction, employing a flow-through cell [Hammond et al. Cryst. Growth Des. 2004, 4, 943] and previously developed chemometric method [Chen Anal. Chem. 2005, 77, 6563], was applied for quantitative analysis of the polymorphic phase transformation from metastable α-form to stable β-form l-glutamic acid (LGA). The process interconversion monitored in aqueous slurries, using jacketed 400 mL magnetically stirred reactor as function temperature. Calibration studies revealed...
With a view to maintaining the validity of multivariate calibration models for chemical processes affected by temperature fluctuations, loading space standardization (LSS) is proposed. Through application LSS, built at temperatures other than those test samples can provide predictions with an accuracy comparable results obtained constant temperature. Compared methods, designed same purpose, such as continuous piecewise direct standardization, LSS has advantages straightforward implementation...