Optimality assessment with optimality recovery for multi-modal process operations
dc.contributor.advisor | Louw, Tobias Muller | en_ZA |
dc.contributor.advisor | Bradshaw, Steven Martin | en_ZA |
dc.contributor.author | Meyer, Tanya | en_ZA |
dc.contributor.other | Stellenbosch University. Faculty of Engineering. Dept. of Process Engineering. | en_ZA |
dc.date.accessioned | 2023-03-03T12:30:49Z | en_ZA |
dc.date.accessioned | 2023-05-18T06:57:30Z | en_ZA |
dc.date.available | 2023-03-03T12:30:49Z | en_ZA |
dc.date.available | 2023-05-18T06:57:30Z | en_ZA |
dc.date.issued | 2023-03 | en_ZA |
dc.description | Thesis (MEng)--Stellenbosch University, 2023. | en_ZA |
dc.description.abstract | ENGLISH ABSTRACT: The field of optimality assessment (OA) is a recent development within data-driven process monitoring. OA is a plant-wide approach to real-time optimisation that aims to minimise nonoptimal online operation caused by (1) disturbances that cannot be rejected by the regulatory control system, or (2) inevitable controlled variable setpoint drift. The distinguishing design factor of OA, as opposed to fault- and quality-related process monitoring, is the incorporation of the comprehensive economic index to quantify overall plant optimality or performance. Since optimality is only available in retrospect, the estimation of optimality during real-time operation allows for prompt intervention when nonoptimal conditions arise, so as to prevent prolonged conditions of deteriorated Performance. This work proposes an alternative to the conventional latent variable model-based OA workflows, which employ monitoring charts founded on Shewhart- or similarity-based statistics. The proposed OA workflow is designed to account for continuous and multimodal industrial process data without transition states. The workflow is developed under the framework of a novel optimality landscape which captures various stable modes in the historical process dataset as well as their associated optimality grade. In addition, the proposed optimality landscape captures the cause for the historical operating point shifting from one mode to another, which is termed a modal shift. Two types of modal shifts are captured, namely those that are caused by disturbances, or by SP change(s) that are implemented by the control system or operational team. The offline phase of the proposed OA workflow constructs a holistic reference tool called the optimality graph. The nodes of the optimality graph are discovered by 𝑘-means clustering in the latent variable space, whereas the edges are discovered by the proposed TASLA (time-based alignment of modal shifts and plant log algorithm) technique. Furthermore, metrics are developed for selecting hyperparameters that result in an optimality graph which best reflects the optimality landscape. The online phase of the proposed workflow essentially projects online conditions onto the historical optimality graph using latent variable techniques, such that the closest reference mode is identified. Consequently, real-time conditions are assigned the optimality grade of the identified reference mode. The online optimality graph reveals which actions can be implemented to shift the online operating point toward modes of differing optimality. A key outcome of this work is the utility of the online optimality graph as an optimality advisory that provides the operational team with historically substantiated decision support for implementing optimality recovery. The performance of the proposed OA workflow is tested using a simulated Tennessee-Eastman Process dataset. The proposed workflow captures the historical optimality landscape of the pseudo-industrial dataset and estimates real-time optimality well upon comparison to the ground truth. The performance of each latent variable extraction technique – namely PCA, PLS, ICA – is evaluated by considering the spread of optimality within each mode of the related optimality graph. PLS is deemed the most suited to extracting features that are reflective of the optimality landscape. The holistic nature of the proposed PLS-based OA workflow offers a good alternative to existing latent variable model-based OA workflows. | en_ZA |
dc.description.abstract | AFRIKAANS OPSOMMING: Die optimaliteitassesseringsveld (OA) is ’n onlangse ontwikkeling binne data-gedrewe prosesmonitering. OA is ’n aanleg-wye benadering tot intydse optimering wat beoog om nie-optimale aanlynbedryf wat veroorsaak word deur (1) steuringe wat nie verwerp kan word deur die regulerende beheersisteem nie, of (2) onvermydelike beheerde veranderlike setpuntbeweging, te minimeer. Die onderskeidende ontwerpfaktor van OA is, teenoor fout- en kwaliteit-verwante prosesmonitering, die inkorporasie van die omvattende ekonomiese indeks om algehele aanlegoptimaliteit en -doeltreffendheid te kwantifiseer. Aangesien optimaliteit slegs in retrospek beskikbaar is, maak die beraming van optimaliteit gedurende intydse bedryf, spoedige ingryping moontlik wanneer nie-optimale kondisies voorkom, sodat verlengde kondisies van verslegtende doeltreffendheid verhoed kan word. Hierdie werk stel ’n alternatief tot die konvensionele latente veranderlike model-gebaseerde OA-werkvloei voor, wat moniteringskaarte gegrond op Shewhart- of soortgelykheid-gebaseerde statistiek gebruik. Die voorgestelde OA-werkvloei is ontwerp om rekening te hou van aaneenlopende en multimodale industriële prosesdata sonder oorgangstoestande. Die werkvloei is ontwikkel onder die raamwerk van ’n nuwe optimaliteitlandskap wat verskeie stabiele modusse in die historiese proses datastel sowel as hul geassosieerde optimaliteitsgraad vasvang. Hierby, die voorgestelde optimaliteitslandskap vang die rede vir die historiese bedryfspunt wat van een modus na ’n ander skuif, wat ’n modale skuif genoem word, vas. Twee tipes modale skuiwe word vasgevang, naamlik dié wat veroorsaak word deur steuringe, of deur SP-verandering(e) wat geïmplementeer word deur die beheersisteem of bedryfspan. Die aflynfase van die voorgestelde OA-werkvloei bou ’n holistiese verwysingsimplement wat die optimaliteitsgrafiek genoem word. Die nodes van die optimaliteitsgrafiek word ontdek deur 𝑘-gemiddeldes groepering in die latente veranderlike spasie, waar teenoor rante ontdek word deur die voorgestelde TASLA (tyd-gebaseerde algoritme) tegniek. Verder, metrieke word ontwikkel vir selektering van hiperparameters wat ’n optimaliteitsgrafiek tot gevolg het wat die optimaliteitslandskap die beste reflekteer. Die aanlynfase van die voorgestelde werkvloei projekteer in wese aanlynkondisies op die historiese optimaliteitsgrafiek deur latente veranderlike tegnieke te gebruik, op so ’n manier dat die naaste verwysingsmodus geïdentifiseer word. Vervolgens, intydse kondisies word die optimaliteitsgraad van die geïdentifiseerde verwysingsmodus toegeken. Die aanlyn-optimaliteitsgrafiek maak aksies wat geïmplementeer kan word bekend om die aanlyn-bedryfspunt na modusse van verskillende optimaliteit te skuif. ’n Sleutel uitkomste van hierdie werk is die gebruik van die aanlyn-optimaliteitsgrafiek as ’n optimaliteitsadviseur wat die bedryfspan met histories bevestigde ondersteuning vir implementering van optimaliteitsherwinning, verskaf. Die doeltreffendheid van die voorgestelde OA-werkvloei is getoets deur ’n gesimuleerde Tennessee-Eastman-prosesdatastel te gebruik. Die voorgestelde werkvloei vang die historiese optimaliteitslandskap van die pseudo-industriële datastel en beraam intydse optimaliteit goed in vergelyking met die grondwaarde. Die doeltreffendheid van elke latente veranderlike ekstraksietegniek – naamlik PCA, PLS, ICA – is geëvalueer deur die verspreiding van optimaliteit binne elke modus van die verwante optimaliteitsgrafiek te oorweeg. PLS is beskou as die mees-gepaste vir ekstraksie-eienskappe wat die optimaliteitslandskap reflekteer. Die holistiese natuur van die voorgestelde PLS-gebaseerde OA-werkvloei bied ’n goeie alternatief vir die bestaande latente veranderlike model-gebaseerde OA-werkvloeie. | af_ZA |
dc.description.version | Masters | en_ZA |
dc.format.extent | xiii, 128 pages : illustrations | en_ZA |
dc.identifier.uri | http://hdl.handle.net/10019.1/126956 | en_ZA |
dc.language.iso | en_ZA | en_ZA |
dc.language.iso | en_ZA | en_ZA |
dc.publisher | Stellenbosch : Stellenbosch University | en_ZA |
dc.rights.holder | Stellenbosch University | en_ZA |
dc.subject.lcsh | Real-time data processing | en_ZA |
dc.subject.lcsh | Production engineering | en_ZA |
dc.subject.lcsh | Optimisation | en_ZA |
dc.subject.lcsh | Process control -- Data processing | en_ZA |
dc.title | Optimality assessment with optimality recovery for multi-modal process operations | en_ZA |
dc.type | Thesis | en_ZA |
Files
Original bundle
1 - 1 of 1
Loading...
- Name:
- meyer_optimality_2023.pdf
- Size:
- 12.45 MB
- Format:
- Adobe Portable Document Format
- Description: