Thesis (MEng) -- Stellenbosch University, 2014. / ENGLISH ABSTRACT: Modern mineral processing plants are characterised by a large number of measured variables, interacting through numerous processing units, control loops and often recycle streams. Consequentially, faults in these plants propagate throughout the system, causing significant degradation in performance. Fault diagnosis therefore forms an essential part of performance monitoring in such processes. The use of feature extraction methods for fault diagnosis has been proven in literature to be useful in application to chemical or minerals processes. However, the ability of these methods to identify the causes of the faults is limited to identifying variables that display symptoms of the fault. Since faults propagate throughout the system, these results can be misleading and further fault identification has to be applied. Faults propagate through the system along material, energy or information flow paths, therefore process topology information can be used to aid fault identification. Topology information can be used to separate the process into multiple blocks to be analysed separately for fault diagnosis; the change in topology caused by fault conditions can be exploited to identify symptom variables; a topology map of the process can be used to trace faults back from their symptoms to possible root causes. The aim of this project, therefore, was to develop a process monitoring strategy that exploits process topology for fault detection and identification. Three methods for extracting topology from historical process data were compared: linear cross-correlation (LC), partial cross-correlation (PC) and transfer entropy (TE). The connectivity graphs obtained from these methods were used to divide process into multiple blocks. Two feature extraction methods were then applied for fault detection: principal components analysis (PCA), a linear method, was compared with kernel PCA (KPCA), a nonlinear method. In addition, three types of monitoring chart methods were compared: Shewhart charts; exponentially weighted moving average (EWMA) charts; and cumulative sum (CUSUM) monitoring charts. Two methods for identifying symptom variables for fault identification were then compared: using contributions of individual variables to the PCA SPE; and considering the change in connectivity. The topology graphs were then used to trace faults to their root causes. It was found that topology information was useful for fault identification in most of the fault scenarios considered. However, the performance was inconsistent, being dependent on the accuracy of the topology extraction. It was also concluded that blocking using topology information substantially improved fault detection and fault identification performance. A recommended fault diagnosis strategy was presented based on the results obtained from application of all the fault diagnosis methods considered. / AFRIKAANSE OPSOMMING: Moderne mineraalprosesseringsaanlegte word gekarakteriseer deur ʼn groot aantal gemete veranderlikes, wat in wisselwerking tree met mekaar deur verskeie proseseenhede, beheerlusse en hersirkulasiestrome. As gevolg hiervan kan foute in aanlegte deur die hele sisteem propageer, wat prosesprestasie kan laat afneem. Foutdiagnose vorm dus ʼn noodsaaklike deel van prestasiemonitering. Volgens literatuur is die gebruik van kenmerkekstraksie metodes vir foutdiagnose nuttig in chemiese en mineraalprosesseringsaanlegte. Die vermoë van hierdie metodes om die fout te kan identifiseer is egter beperk tot die identifikasie van veranderlikes wat simptome van die fout vertoon. Aangesien foute deur die sisteem propageer kan resultate misleidend wees, en moet verdere foutidentifikasie metodes dus toegepas word. Foute propageer deur die proses deur materiaal-, energie- of inligtingvloeipaaie, daarom kan prosestopologie inligting gebruik word om foutidentifikasie te steun. Topologie inligting kan gebruik word om die proses in veelvoudige blokke te skei om die blokke apart te ontleed. Die verandering in topologie veroorsaak deur fouttoestande kan dan analiseer word om simptoomveranderlikes te identifiseer. ʼn Topologiekaart van die proses kan ontleed word om moontlike hoofoorsake van foute op te spoor. Die doel van hierdie projek was dus om ʼn prosesmoniteringstrategie te ontwikkel wat prosestopologie benut vir fout-opspooring en foutidentifikasie. Drie metodes vir topologie-ekstraksie van historiese prosesdata is met mekaar vergelyk: liniêre kruiskorrelasie, parsiële kruiskorrelasie en oordrag-entropie. Konnektiwiteitsgrafieke verkry deur hierdie ekstraksie-metodes is gebruik om die proses in veelvoudige blokke te skei. Twee kenmerkekstraksiemetodes is hierna toegepas om foutdeteksie te bewerkstellig: hoofkomponentanalise (HKA), ʼn liniêre metode; en kernhoofkomponentanalise (KHKA), ʼn nie-lineêre metode. Boonop was drie tipes moniteringskaart metodes vergelyk: Shewhart kaarte, eksponensieel-geweegde bewegende gemiddelde kaarte en kumulatiewe som kaarte. Twee metodes om simptoom veranderlikes te identifiseer vir foutidentifikasie was daarna vergelyk: gebruik van individuele veranderlikes; en inagneming van die verandering in konnektiwiteit. Die konnektiwiteitgrafieke was daarna gebruik om hoofoorsake van foute op te spoor. Dit is gevind dat topologie informasie nuttig was vir foutidentifikasie vir meeste van die fouttoestande ondersoek. Nogtans was die prestasie onsamehangend, aangesien dit afhanklik is van die akkuraatheid waarmee topologie ekstraksie uitgevoer is. Daar was ook afgelei dat die gebruik van topologie blokke beduidend die fout-opspooring en foutidentifikasie prestasie verbeter het. ʼn Aanbevole foutdiagnose strategie is voorgestel.

Regenerative medicine (RM) represents a promising enabling technology to revolutionize healthcare. This said there are still major gaps between the commercial promise and the reality of the cell therapy sector of regenerative medicine. There is consensus to develop high through-put, automated technologies for the manufacture of RM products. Imaging methods will have the capacity to contribute to this technological gap for cell therapies and are particularly attractive to provide non-destructive monitoring with high spatial and temporal resolution. This work applied an automated, non-invasive phase contrast imaging platform (Cell-IQ) to measure, analyse and ultimately quantify image derived metrics for human embryonic stem cells (hESCs) and haematopoietic stem cells (HSCs) as part of the colony forming unit (CFU) assay. This work has shown through thresholding and machine vision identification technology, imaging has the ability to improve the precision of current evaluation methods for cell culture, providing novel information regarding culture state and show image derived metrics to be predictive of future culture state. Building on this, differentiation through the addition of a growth factor cocktail highlighted how in-process monitoring enables protocol optimisation. After equilibrating the Cell-IQ incubator to a standard incubator, the progress of the CFU assay was monitored and image metrics representative of colony phenotype were analysed. Cell count, distance between cells and cell migration within individual colonies were identified to be informative and provide a degree of colony phenotype separation. Quantitative, novel, image derived metrics were identified that improve reliability through computer automation, cost by removing user verification and time by reducing the assay time from 14 days to 7 days. Non-invasive imaging provides a fantastic opportunity to create bespoke sampling frequencies to achieve desired precision for manufacturing cell therapies, this work has developed and shown improvement and a level of control to current culture process for ESCs and HSCs.

Thesis (MEng)-- Stellenbosch University, 2014. / ENGLISH ABSTRACT: The South African industry has an increasing need for manufacturing automation. However, the classical form of automation is not cost effective for the low volumes and high variance of products that are produced there. The industry may use the reconfigurable manufacturing system (RMS) concept to improve production of its products. However, industry has been unwilling to adopt the reconfigurable manufacturing systems developed in recent research projects. Due to industry’s hesitance to adopt the control platforms on which reconfigurable manufacturing systems are currently based, the focus of the thesis is on creating a reconfigurable control system using industry accepted technologies. This research focused on evaluating a Beckhoff embedded PC’s suitability as a station controller that controls a reconfigurable subsystem in an RMS. The control system for the station controller was developed using only the IEC 61131-3 programming languages and the Beckhoff programming software. This control system was evaluated by using it to control a station that is responsible for testing a circuit breaker’s tripping current and time. The developed control system was based on the ADACOR architecture because of its optimisation capabilities that were necessary to keep the cycle time of the station as low as possible. The design and implementation of the physical configuration and control system of the station is described in this thesis. The station was designed to meet the requirements of both an RMS and the case study. Because of the limitations of the IEC 61131-3 programming languages, dynamic instantiation of holons is not possible and a method was developed to simulate dynamic task holons. By making use of the embedded PC’s ability to run multiple PLCs at the same time, each type of holon was run in its own PLC thread. The developed control system and station was evaluated by conducting experiments using a laboratory test setup. The evaluation of the developed control system in this thesis proved that an RMS can be created, in the context of station control, using IEC 61131-3 and industry accepted technologies, if a hardware platform is used that allows multiple PLCs to be run in individual threads. The control approach that was created in this thesis can be used to create station control systems that offers optimised cycle times, the benefits of an RMS and the benefits of industry accepted technology. / AFRIKAANSE OPSOMMING: Die Suid-Afrikaanse bedryf het 'n toenemende behoefte aan geoutomatiseerde vervaardiging. Die klassieke vorm van outomatisasie is egter nie koste effektief vir die lae volumes en hoë variansie van produkte wat in Suid Afrika geproduseer word nie. Die bedryf kan moontlik die konsep van 'n herkonfigureerbare vervaardigingstelsel (HVS) gebruik om vervaardiging te outomatiseer. Die bedryf is egter nie bereid om die herkonfigureerbare vervaardigingstelsels wat in onlangse navorsingsprojekte ontwikkel is, te aanvaar nie. As gevolg van die bedryf se huiwering om die beheerplatforms waarop herkonfigureerbare vervaardigingstelsels tans gebaseer word, te aanvaar, is die fokus van die tesis om industrie-aanvaarde tegnologie te gebruik om ‘n herkonfigureerbare beheerstelsel te skep. Hierdie navorsing fokus op die evaluering van 'n “Beckhoff embedded PC” se geskiktheid as 'n stasiebeheerder van 'n herkonfigureerbare substelsel in 'n HVS. Die beheerstelsel vir die stasie beheerder is ontwikkel deur slegs van die IEC 61131-3 programmeringstale en die Beckhoff programmering-sagteware gebruik te maak. Hierdie beheerstelsel is geëvalueer deur dit op die beheer van 'n stasie wat verantwoordelik is vir die toets stroombrekers, toe te pas. Die beheerstelsel was gebaseer op die ADACOR argitektuur as gevolg van die optimeringsvermoëns wat noodsaaklik was om die siklustyd van die stasie so laag as moontlik te hou. Die ontwerp en implementering van die fisiese konfigurasie en beheerstelsel van die stasie word in hierdie tesis beskryf. Die stasie was ontwerp om aan die vereistes van beide 'n HVS en die gevallestudie te voldoen. As gevolg van die beperkings van die IEC 61131-3 programmeringstale, is dinamiese instansiëring van holons nie moontlik nie, en 'n metode is ontwikkelom dinamiese taakholons na te boots. Deur gebruik te maak van die "embedded PC" se vermoë om meervoudige PLCs terselfdetyd te hanteer, kan elke holon tipe in sy eie "thread" loop. Die ontwikkelde stelsel en die stasie is geëvalueer in 'n laboratorium deur middel van eksperimente. Die evaluering van die beheerstelsel in hierdie tesis bewys dat 'n HVS geskep kan word, in die konteks van ‘n stasiebeheerder, deur IEC 61131-3 en tegnologie wat wyd in die industrie aanvaar word, te gebruik mits die hardeware-platform wat gebruik word toelaat dat verskeie PLCs terselfde tyd op een beheerder kan loop. Die beheerbenadering wat geskep is in hierdie tesis kan gebruik word om stasie- beheerstelsels te skep wat optimale siklus tye, die voordele van 'n HVS en die voordele van industrie-aanvaarde tegnologie bied.

Thesis (MScIng)--University of Stellenbosch, 2003. / ENGLISH ABSTRACT: The analysis of process data obtained from chemical and metallurgical engineering systems is a crucial aspect of the operating of any process, as information extracted from the data is used for control purposes, decision making and forecasting. Singular spectrum analysis (SSA) is a relatively new technique that can be used to decompose time series into their constituent components, after which a variety of further analyses can be applied to the data. The objectives of this study were to investigate the abilities of SSA regarding the filtering of data and the subsequent modelling of the filtered data, to explore the methods available to perform nonlinear SSA and finally to explore the possibilities of Monte Carlo SSA to characterize and identify process systems from observed time series data. Although the literature indicated the widespread application of SSA in other research fields, no previous application of singular spectrum analysis to time series obtained from chemical engineering processes could be found. SSA appeared to have a multitude of applications that could be of great benefit in the analysis of data from process systems. The first indication of this was in the filtering or noise-removal abilities of SSA. A number of case studies were filtered by various techniques related to SSA, after which a number of neural network modelling strategies were applied to the data. It was consistently found that the models built on data that have been prefiltered with SSA outperformed the other models. The effectiveness of localized SSA and auto-associative neural networks in performing nonlinear SSA were compared. Both techniques succeeded in extracting a number of nonlinear components from the data that could not be identified from linear SSA. However, it was found that localized SSA was a more reliable approach, as the auto-associative neural networks would not train for some of the data or extracted nonsensical components for other series. Lastly a number of time series were analysed using Monte Carlo SSA. It was found that, as is the case with all other characterization techniques, Monte Carlo SSA could not succeed in correctly classifying all the series investigated. For this reason several tests were used for the classification of the real process data. In the light of these findings, it was concluded that singular spectrum analysis could be a valuable tool in the analysis of chemical and metallurgical process data. / AFRIKAANSE OPSOMMING: Die analise van chemise en metallurgiese prosesdata wat verkry is vanaf chemiese of metallurgiese ingenieursstelsels is ‘n baie belangrike aspek in die bedryf van enige proses, aangesien die inligting wat van die data onttrek word vir prosesbeheer, besluitneming of die bou van prosesmodelle gebruik kan word. Singuliere spektrale analise is ‘n relatief nuwe tegniek wat gebruik kan word om tydreekse in hul onderliggende komponente te ontbind. Die doelwitte van hierdie studie was om ‘n omvattende literatuuroorsig oor die ontwikkeling van die tegniek en die toepassing daarvan te doen, beide in die ingenieursindustrie en in ander navorsingsvelde, die navors van die moontlikhede van SSA aangaande die verwydering van geraas uit die data en die gevolglike modellering van die skoon data te ondersoek, ‘n ondersoek te doen na sommige van die beskikbare tegnieke vir nie-lineêre SSA en laastens ‘n studie te maak van die potensiaal van Monte Carlo SSA vir die karakterisering en identifikasie van data verkry vanaf prosesstelsels. Ten spyte van aanduidings in die literatuur dat SSA wydverspreid toegepas word in ander navorsingsvelde, kon geen vorige toepassings gevind word van SSA op chemiese prosesse nie. Dit wil voorkom asof die chemiese nywerhede groot baat kan vind by SSA van prosesdata. Die eerste aanduiding van hierdie voordele was in die vermoë van SSA om geraas te verwyder uit tydreekse. ‘n Aantal tipiese gevalle is ondersoek deur van verskeie benaderings tot SSA gebruik te maak. Nadat die geraas uit die tydreekse van die toetsgevalle verwyder is, is neurale netwerke gebruik om die prosesse te modelleer. Daar is herhaaldelik gevind dat die modelle wat gebou is op data wat eers deur SSA skoongemaak is, beter presteer as die wat slegs op die onverwerkte data gepas is. Die effektiwiteit van lokale SSA en auto-assosiatiewe neurale netwerke om nie- lineêre SSA toe te pas is ook vergelyk. Albei tegnieke het daarin geslaag om nie- lineêre hoofkomponente van die data te onttrek wat nie geïdentifiseer kon word deur die lineêre benadering nie. Daar is egter gevind dat lokale SSA ‘n meer betroubare tegniek is, aangesien die autoassosiatiewe neurale netwerke nie op sommige van die datastelle wou leer nie en vir ander tydreekse sinnelose hoofkomponente onttrek het. Laastens is ‘n aantal tydreekse geanaliseer met behulp van Monte Carlo SSA. Soos met alle ander karakteriseringstegnieke, kon Monte Carlo SSA nie daarin slaag om al die tydreekse wat ondersoek is korrek te identifiseer nie. Om hierdie rede is ‘n kombinasie van toetse gebruik om die onbekende tydreekse te klassifiseer. In die lig van al hierdie bevindinge, is die gevolgtrekking gemaak dat singuliere spektrale analise ‘n waardevolle hulpmiddel kan wees in die analise van chemiese en metallurgiese prosesdata.

Thesis (MScIng)--University of Stellenbosch, 2003. / ENGLISH ABSTRACT: Physical systems encountered in process engineering are invariably ill-defined, multivariate, and exhibit complex nonlinear dynamical behaviour. The increasing demands for better process efficiency and high product quality have led to the development and implementation of advanced control strategies in process plants. These modern control strategies are based on the use of a mathematical model defined for the process. Traditionally, linear models have been used to approximate the dynamics of processes whereas most processes are governed by nonlinear mechanisms. Since linear systems theory is well-established whereas nonlinear systems theory is not, recent developments in nonlinear dynamical systems theory present opportunities for improved approaches in modelling these process systems. It is now known that a nonlinear description of a process can be obtained from using time-delayed copies reconstructed from measurements taken from the process. Due to low signal to noise ratios associated with measured data it is logical to exploit redundant information in multivariate time signals taken from the systems in reconstructing the underlying dynamics. This study investigated the extension of univariate nonlinear time series analysis to the situation where multivariate measurements are available. Using simulated data from a coupled continuously stirred tank reactor and measured data from a flotation process system, the comparative advantages of using multivariate and univariate state space reconstructions were investigated. With respect to detection of nonlinearity multivariate surrogate analysis were found to give potentially robust results because of preservation of cross-correlations among components in the surrogate data. Multivariate local linear models showed a deterministic structure in both small and large neighbourhood sizes whereas for scalar embeddings determinism was defined only in smaller neighbourhood sizes. Non-uniform multivariate embeddings gave local linear models that resembled models from a trivial reconstruction of the original state space variables. With regard to global nonlinear modelling, multivariate embeddings gave models with better predictability irrespective of the model class used. Further improvements in the performance of models were obtained for multivariate non-uniform embeddings. A relatively new statistical learning algorithm, the least-squares support vector machine (LSSVM), was evaluated using multilayer perceptrons (MLP) as a benchmark in modelling nonlinear time series using simulated and plant data. It was observed that in the absence of autocorrelations in the variables and sparse data LSSVMs performed better than MLPs. Simulation of trained models gave consistent results for the LSSVMs, which was not the case for MLPs. However, the computational costs incurred in training the LSSVM model was significantly higher than for MLPs. LSSVMs were found to be insensitive to dimensionality reduction methods whereas the performance of MLPs degraded with increasing complexity of the dimension reduction method. No relative merits were found for using complex subspace dimension reduction methods for the data used. No general conclusions could be drawn with respect to the relative superiority of one class of models method over the other. Spatiotemporal structures are routinely observed in many chemical systems, such as reactive-diffusion and other pattern forming systems. We investigated the modelling of spatiotemporal time series using the coupled logistic map lattice as a case study. It was found that including both spatial and temporal information improved the performance of the fitted models. However, the superiority of spatiotemporal embeddings over individual time series was found to be defined for certain choices of the spatial and temporal embedding parameters. / AFRIKAANSE OPSOMMING: Fisiese stelsels wat in prosesingenieurswese voorkom is dikwels nie goed gedefinieer nie, multiveranderlik en vertoon komplekse nie-lineˆere gedrag. Toenemende vereistes vir ho¨e prosesdoeltreffendheid en produkgehalte het gelei tot die ontwikkeling en implementering van gevorderde beheerstrategie¨e vir prosesaanlegte. Hierdie morderne beheerstrategie¨e is gebaseer op die gebruik van wiskundige prosesmodelle. Lineˆere modelle word gewoonlik ontwikkel, al is die onderliggende prosesmeganismes in die algemeen nie-lineˆere, aangesien lineˆere stetselteorie goed gevestig is, en nie-line¨ere stelselteorie nie. Onlangse verwikkelinge in die teorie van nie-lineˆeredinamiese stelsels bied egter geleenthede vir verbeterde modellering van prosesstelsels. Dit is bekend dat ‘n nie-lineˆere beskrywing van ‘n progses verkry kan word deur tydvertraagde kopie¨e van metings van die prosesse te rekonstrueer. Met die lae seintot- geraasverhoudings wat met gemete data geassosieer word, is dit logies om die oortollige informasie in meerveranderlike seine te benut tydens die rekonstruksie van die onderliggende prosesdinamika. In die tesis is die uitbreiding van enkel-veranderlike nie-lineˆere tydreeksontleding na meer-veranderlike stelsels ondersoek. Met data van twee aaneengeskakelde gesimuleerde geroerde tenkreaktore en werklike data van ‘n flottasieproses, is die meriete van enkel- en meerveranderlike rekonstruksies van toestandruimtes ondersoek. Meerveranderlike surrogaatdata-ontleding het nie-lineariteite in die data op ‘n meer robuuste wyse ge¨ıdentifiseer, a.g.v. die behoud van kruis-korrelasies in die komponente van die data. Meerveranderlike lokale lineˆere modelle het ‘n deterministiese struktuur in beide klein en groot naasliggende omgewings ge¨ıdentifiseer, terwyl enkelveranderlike metodes dit slegs vir klein naasliggende omgewings kon doen. Nie-uniforme meerveranderlike inbeddings het lokale lineˆere modelle gegenereer wat soos globale modelle afkomstig van triviale rekonstruksies van die data gelyk het. M.b.t globale nie-lineˆere modellering, het meerveranderlike inbedding deurgaans beter modelle opgelewer. Verdere verbetering in die prestasie van modelle kon verkry word d.m.v. meerveranderlike nie-uniforme inbedding. ‘n Relatief nuwe statistiese algoritme, die kleinste-kwadrate-steunvektormasjien (KKSVM) is ge¨evalueer teenoor multilaag-perseptrons (MLP) as ‘n standaard vir die modellering van nie-lineˆere tydreekse, deur gebruik te maak van gesimuleerde en werklike aanlegdata. Daar is gevind dat die KKSVM beter presteer het as die MLPs wanneer die opeenvolgende waarnemings swak gekorreleer en min was relatief tot die aantal veranderlikes. Die KKSVMs het beduidend langer geneem as die MLPs om te ontwikkel. Hulle was ook minder sensitief vir die metodes wat gevolg is om die dimensionaliteit van die data te verlaag, anders as die MLPs. Ook is gevind dat meer komplekse metodes tot die verlaging van die dimensionaliteit weinig nut gehad het. Geen algemene gevolgtrekkings kan egter gemaak word m.b.t die verskillende modelle nie. Ruimtelik-temporale strukture word algemeen waargeneem in baie chemiese stelsels, soos reaktiewe diffusie e.a. patroonvormende sisteme. Die modellering van ruimtelik-temporale stelsels is bestudeer aan die hand van ‘n gekoppelde logistiese projeksierooster. Insluiting van beide die ruimtelike en temporale inligting het tot beduidend beter modelle gelei, solank as wat di´e inligting op die regte wyse ontsluit is.

Model predictive control has become a standard for most control strategies in modern process plants. It relies heavily on process models, which might not always be fundamentally available, but can be obtained from time series analysis. The first step in any control strategy is to identify or detect changes in the system, if present. The detection of such changes, known as dynamic changes, is the main objective of this study. In the literature a wide range of change detection methods has been developed and documented. Most of these methods assume some prior knowledge of the system, which is not the case in this study. Furthermore a large number of change detection methods based on process history data assume a linear relationship between process variables with some stochastic influence from the environment. These methods are well developed, but fail when applied to nonlinear dynamic systems, which is focused on in this study. A large number of the methods designed for nonlinear systems make use of statistics defined in phase space, which led to the method proposed in this study. The correlation dimension is an invariant measure defined in phase space that is sensitive to dynamic change in the system. The proposed method uses the correlation dimension as test statistic with and moving window approach to detect dynamic changes in nonlinear systems. The proposed method together with two dynamic change detection methods with different approaches was applied to simulated time series data. The first method considered was a change-point algorithm that is based on singular spectrum analysis. The second method applied to the data was mutual cross prediction, which utilises the prediction error from a multilayer perceptron network. After the proposed method was applied to the data the three methods’ performance were evaluated. Time series data were obtained from simulating three systems with mathematical equations and observing one real process, the electrochemical noise produced by a corroding system. The three simulated systems considered in this study are the Belousov-Zhabotinsky reaction, an autocatalytic process and a predatory-prey model. The time series obtained from observing a single variable was considered as the only information available from the systems. Before the change detection methods were applied to the time series data the phase spaces of the systems were reconstructed with time delay embedding. All three the methods were able to do identify the change in dynamics of the time series data. The change-point detection algorithm did however produce a haphazard behaviour of its detection statistic, which led to multiple false alarms being encountered. This behaviour was probably due to the distribution of the time series data not being normal. The haphazard behaviour reduces the ability of the method to detect changes, which is aggravated by the presence of chaos and instrumental or measurement noise. Mutual cross prediction is a very successful method of detecting dynamic changes and is quite robust against measurement noise. It did however require the training of a multilayer perceptron network and additional calculations that were time consuming. The proposed algorithm using the correlation dimension as test statistic with a moving window approach is very useful in detecting dynamic changes. It produced the best results on the systems considered in this study with quick and reliable detection of dynamic changes, even in then presence of some instrumental noise. The proposed method with the correlation dimension as test statistic was the only method applied to the real time series data. Here the method was successful in distinguishing between two different corrosion phenomena. The proposed method with the correlation dimension as test statistic appears to be a promising approach to the detection of dynamic change in nonlinear systems.

An artificial neural network is a mathematical black-box modelling tool. This tool can be used to model complex non-linear multivariable processes. In attempting to create an inverse process model of an industrial linear low density polyethylene reactor, several interesting results were encountered. Both time-invariant algebraic and time-invariant dynamic models could adequately represent the process, provided an identified 50-minute time lag was taken into account. A novel variation of the traditional IMC controller was implemented which used two inverse neural network process models. This was named Inverse Internal Model Control (IIMC). This controller was initially tested on a real multivariable pump-tank system and showed promising results. The IIMC controller was adapted to an on-line version for the polymer plant control system. The controller was run in open loop mode to compare the predictions of the controller with the actual PID ratio controllers. It was hoped that by incorporating neural network models into the controller, they would take the non-linearity and coupling of the variables into account, which the present PID controllers are unable to do. The existing PID controllers operate on separate loops involving the two main feeds (co-monomer and hydrogen) to the reactor, which constitute aspects of the control system in which the scope for advanced control exists. Although the control loop was not closed, the groundwork has been laid to implement a novel controller that could the operation of the plant. / Thesis (M.Sc.Eng.)-University of Natal, Durban, 2001.

Dynamic Matrix Control (DMC) has proven to be a powerful tool for optimal regulation of chemical processes under constrained conditions. The internal model of this predictive controller is based on step response measurements at an average operating point. As the process moves away from this point, however, control becomes sub-optimal due to process non-linearity. If DMC is made adaptive, it can be expected to perform well even in the presence of uncertainties, non-linearities and time-vary ing process parameters. This project examines modelling and control issues for a complex multivariable industrial operator training plant, and develops and applies a method for adapting the controller on-line to account for non-linearity. A two-input/two-output sub-system of the Training Plant was considered. A special technique had to be developed to deal with the integrating nature of this system - that is, its production of ramp outputs for step inputs. The project included the commissioning of the process equipment and the addition of instrumentation and interfacing to a SCADA system which has been developed in the School of Chemical Engineering. / Thesis (M.Sc.Eng.)-University of Natal, Durban, 2001.

Submitted in fulfillment of the requirements for the Master of Engineering Degree, Durban University of Technology, Durban, South Africa, 2016. / This research discusses the design and development of a computational intelligent based diagnostic system to assess the operating state of a process control valve. Process control valves react to a controller signal and are the main source of faults in a control loop. The elasticity inherent within a valve’s mechanical construction makes it prone to nonlinearities such as backlash, hysteresis and stiction. These nonlinearities negatively affect the performance of a process control loop during a control session. The diagnostic system proposed in this research utilises artificial neural network systems configured as ensembles to classify common control valve faults. Each ensemble functions as a ‘specialist’ trained to identify a specific loop fault. The team of specialized artificial neural networks are configured into a single comprehensive system to detect common control loops problems such as valve hysteresis, backlash, stiction and low air supply. The detection of a specific type of fault is achieved by comparing the mean square error output from each network. The ensemble having the lowest mean square error is the network that has been trained to identify a specific type of fault. Two practical methods to simulate control valve stiction and hysteresis are also presented in this study. These methods make it possible for researchers to investigate dynamics of nonlinear behaviour when these nonlinear effects occur in the control channel. / M

No description available.