ITEM VIEW

Neurocontrol of a multi-effect batch distillation pilot plant based on evolutionary reinforcement learning

dc.contributor.authorConradie A.v.E.
dc.contributor.authorAldrich C.
dc.date.accessioned2011-05-15T15:59:00Z
dc.date.available2011-05-15T15:59:00Z
dc.date.issued2010
dc.identifier.citationChemical Engineering Science
dc.identifier.citation65
dc.identifier.citation5
dc.identifier.issn92509
dc.identifier.other10.1016/j.ces.2009.11.003
dc.identifier.urihttp://hdl.handle.net/10019.1/10948
dc.description.abstractThe time cost of first-principles dynamic modelling and the complexity of nonlinear control strategies may limit successful implementation of advanced process control. The maximum return on fixed capital within the processing industries is thus compromised. This study introduces a neurocontrol methodology that uses partial system identification and symbiotic memetic neuro-evolution (SMNE) for the development of neurocontrollers. Partial system identification is achieved using singular spectrum analysis (SSA) to extract state variables from time series data. The SMNE algorithm uses a symbiotic evolutionary algorithm and particle swarm optimisation to learn optimal neurocontroller weights from the partially identified system within a reinforcement learning framework. A multi-effect batch distillation (MEBAD) pilot plant was constructed to demonstrate the real world application of the neurocontrol methodology, motivated by the nonsteady state operation and nonlinear process interaction between multiple distillation columns. Multi-loop proportional integral (PI) control was implemented as a reduced model, reflecting an approach involving no modelling or significant controller tuning. Rapid multiple input multiple out nonlinear controller development was achieved using SSA and the SMNE algorithm, demonstrating comparable time and cost to implementation in relation to the reduced model. The optimal neurocontroller reduced the batch time and therefore the energy consumption by 45% compared to conventional multi-loop SISO PI control. © 2009 Elsevier Ltd. All rights reserved.
dc.subjectAdvanced Process Control
dc.subjectBatch distillation
dc.subjectBatch time
dc.subjectController tuning
dc.subjectDynamic modelling
dc.subjectEnergy consumption
dc.subjectEvolutionary programming
dc.subjectFirst-principles
dc.subjectMathematical modelling
dc.subjectMemetic
dc.subjectMulti-effect
dc.subjectMultiple input multiple out
dc.subjectNeuro control
dc.subjectNeuro controllers
dc.subjectNon linear control
dc.subjectNon-linear controllers
dc.subjectNon-linear dynamics
dc.subjectNonlinear process
dc.subjectNonsteady state
dc.subjectPartial systems
dc.subjectParticle swarm optimisation
dc.subjectPI control
dc.subjectProcessing industry
dc.subjectProportional-integral control
dc.subjectReal-world application
dc.subjectReduced model
dc.subjectSingular spectrum analysis
dc.subjectState variables
dc.subjectSymbiotic evolutionary algorithm
dc.subjectTime cost
dc.subjectTime-series data
dc.subjectComputer programming
dc.subjectComputer simulation
dc.subjectControllers
dc.subjectCost reduction
dc.subjectDistillation
dc.subjectDistillation columns
dc.subjectDynamics
dc.subjectElectric load forecasting
dc.subjectEvolutionary algorithms
dc.subjectIdentification (control systems)
dc.subjectIntelligent control
dc.subjectLearning algorithms
dc.subjectMathematical models
dc.subjectMathematical programming
dc.subjectNeural networks
dc.subjectPilot plants
dc.subjectReinforcement
dc.subjectReinforcement learning
dc.subjectSpectrum analysis
dc.subjectSpectrum analyzers
dc.subjectTime series
dc.subjectTime series analysis
dc.subjectTransients
dc.subjectTwo term control systems
dc.subjectProcess control
dc.titleNeurocontrol of a multi-effect batch distillation pilot plant based on evolutionary reinforcement learning
dc.typeArticle
dc.description.versionArticle


Files in this item

FilesSizeFormatView

There are no files associated with this item.

This item appears in the following Collection(s)

ITEM VIEW