MétaCan
Menu
Back to cohort
Record W4309778900 · doi:10.1016/j.fuel.2022.126715

Mining the synergistic effect in hydrothermal co-liquefaction of real feedstocks through machine learning approaches

2022· article· en· W4309778900 on OpenAlexaff
Jie Yu, Xiaomei Zhong, Zhilin Huang, Xiaoyu Lin, Haiyong Weng, Dapeng Ye, Quan He, Jie Yang

Bibliographic record

VenueFuel · 2022
Typearticle
Languageen
FieldEngineering
TopicThermochemical Biomass Conversion Processes
Canadian institutionsDalhousie University
FundersDepartment of Education, Fujian ProvinceMinjiang UniversityNatural Science Foundation of Fujian Province
KeywordsHydrothermal liquefactionAdaBoostRaw materialYield (engineering)LiquefactionMachine learningArtificial intelligenceComputer scienceBoosting (machine learning)Hydrothermal circulationGradient boostingProcess engineeringChemistryMathematicsRandom forestEnvironmental scienceAlgorithmMaterials scienceChemical engineeringSupport vector machineBiofuelWaste managementEngineeringOrganic chemistryMetallurgy

Abstract

fetched live from OpenAlex

Hydrothermal co-liquefaction (co-HTL) of different feedstocks has received much research attention, not only because its significant importance in real industrial applications, but also due to the potential synergy in biocrude yield by tuning mixed feedstock’s biochemical composition and reaction conditions. Although some attempts have been made to search for the synergy from co-liquefying various feedstocks, these processes were remarkably time and labor consuming, and often with low rate of success. Therefore, this study for the first time employed machine learning algorithms to mine the synergistic effect in co-HTL. Started with single task prediction, three machine learning algorithms, including Adaboost, Gradient Boosting Regression and Random Forest, were trained and tested for predicting co-HTL biocrude yield and relative co-liquefaction effect (CE). It was found that their prediction performances were favorable over traditional mathematical equations, in which Gradient Boosting Regression exhibited the best performance for co-HTL biocrude yield prediction (training and testing R 2 of 0.976 and 0.812 respectively), and Adaboost better estimated relative CE. Feature importance analysis further revealed that co-HTL biocrude yield was mainly influenced by the reaction temperature, but relative CE was closely related to mixed feedstock’s lipid and carbohydrate content, implying that the synergism/antagonism from co-HTL was more dependent on the biochemical composition of mixed feedstock than reaction conditions. Multitask predictions, estimating biocrude yield and relative CE simultaneously that are usually required in real co-HTL practices, suggested Adaboost was the most satisfying algorithm (training R 2 of 0.922) among studied ones. An optimal relative CE of 22.07 % along with 36.31 wt% (daf) biocrude yield could be obtained when the mixed feedstock contained 42.93 % protein, 50.49 % carbohydrate, 6.58 % lipid at a temperature of 320 °C, which were in well agreement with experimental results from co-HTL of biomass model components. A mini application software (exe. file including machine learning algorithm) was also developed for quick estimation of synergy and co-HTL biocrude yield by simply inputting mixed feedstock’s biochemical composition and reaction conditions, showing promising potential for academic and industrial practices to mine the co-HTL synergy and design processes efficiently.

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.

How this classification was reachedexpand

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: Bench or experimental · Consensus signal: Bench or experimental
GenreCandidate signal: Empirical · Consensus signal: Empirical
Teacher disagreement score0.113
Threshold uncertainty score0.381

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 head0.016
GPT teacher head0.218
Teacher spread0.202 · 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

Classification

machine, unvalidated

Machine predicted; a candidate call from one teacher head, not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designBench or experimental
Domainnot available
GenreEmpirical

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations23
Published2022
Admission routes1
Has abstractyes

Explore more

Same venueFuelSame topicThermochemical Biomass Conversion ProcessesFrench-language works237,207