MétaCan
Menu
Back to cohort
Record W4402067153 · doi:10.1002/aic.18584

A reinforcement learning approach with masked agents for chemical process flowsheet design

2024· article· en· W4402067153 on OpenAlex

Why this work is in the frame

A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.

affAt least one author lists a Canadian institution in the pinned OpenAlex snapshot.
fundA Canadian funder is recorded on the work.

Bibliographic record

VenueAIChE Journal · 2024
Typearticle
Languageen
FieldEngineering
TopicScheduling and Optimization Algorithms
Canadian institutionsUniversity of Waterloo
FundersNatural Sciences and Engineering Research Council of Canada
KeywordsReinforcement learningProcess (computing)ReinforcementProcess engineeringComputer scienceBiochemical engineeringEngineeringMaterials scienceArtificial intelligenceComposite materialProgramming language

Abstract

fetched live from OpenAlex

Abstract This study introduces two novel Reinforcement Learning (RL) agents for the design and optimization of chemical process flowsheets (CPFs): a discrete masked Proximal Policy Optimization (mPPO) and a hybrid masked Proximal Policy Optimization (mHPPO). The novelty of this work lies in the use of masking within the hybrid framework, i.e., the incorporation of expert input or design rules that allows the exclusion of actions from the agent's decision spectrum. This work distinguishes from others by seamlessly integrating masked agents with rigorous unit operations (UOs) models, that is, advanced thermodynamic and conservation balance equations, in its simulation environment to design and optimize CPF. The efficacy of these agents, along with performance comparisons, is evaluated through case studies, including one that employs a chemical engineering simulator such as ASPEN Plus®. The results of these case studies reveal learning on the part of the agents, that is, the agent is able to find viable flowsheet designs that meet the stipulated process flowsheet design requirements, for example, achieve a user‐defined product quality.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

Full frame distilled prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.

metaresearch head score (Codex)0.000
metaresearch head score (Gemma)0.000
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: Simulation or modeling
GenreCandidate signal: Methods · Consensus signal: none
Teacher disagreement score0.598
Threshold uncertainty score0.393

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0000.000
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0000.000
Science and technology studies0.0000.000
Scholarly communication0.0000.000
Open science0.0000.000
Research integrity0.0000.000
Insufficient payload (model declined to judge)0.0000.000

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.022
GPT teacher head0.249
Teacher spread0.227 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it