首页|Data from Swiss Federal Institute of Technology Lausanne (EPFL) Provide New Insi ghts into Machine Learning (Accelerated Design of Nickel-cobalt Based Catalysts for Co2 Hydrogenation With Humanin-the-loop Active Machine Learning)
Data from Swiss Federal Institute of Technology Lausanne (EPFL) Provide New Insi ghts into Machine Learning (Accelerated Design of Nickel-cobalt Based Catalysts for Co2 Hydrogenation With Humanin-the-loop Active Machine Learning)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-Data detailed on Machine Learning have been presented. According to news reporting from Sion, Switzerland, by NewsRx j ournalists, research stated, "Thermo-catalytic conversion of CO2 into more valua ble compounds, such as methane, is an attractive strategy for energy storage in chemical bonds and creating a carbon-based circular economy. However, designing heterogeneous catalysts remains a challenging, time- and resource-consuming task ." Funders for this research include FWO, EPFL, Centre Interdisciplinaire de Micros copie Electronique (CIME), FWO, European Research Council (ERC). The news correspondents obtained a quote from the research from the Swiss Federa l Institute of Technology Lausanne (EPFL), "Herein, we present an interpretable, human-in-the-loop active machine learning framework to efficiently plan catalyt ic experiments, execute them in an automated set-up, and estimate the effect of experimental variables on the catalytic activity. A dataset with 48 catalytic ac tivity tests was compiled from a design space of Ni-Co/Al2O3 catalysts with over 50 million potential combinations in only eight iterations. This small dataset was found sufficient to predict CO2 conversion, methane selectivity, and methane space-time yield with remarkable accuracy (R-2 > 0.9) f or untested catalysts and reaction conditions. New experiments and catalysts wer e selected with this methodology, leading to experimental conditions that improv ed the methane space-time yield by nearly 50% in comparison to the previously obtained maximum in the dataset. Interpretation of the model predict ions unveiled the effect of each catalyst descriptor and reaction condition on t he outcome. Particularly, the strong predicted inverse trend between the calcina tion temperature and the catalytic activity was validated experimentally, and ch aracterization implied an underlying structure-performance relationship. Finally,it is demonstrated that the deployed active learning model is excellently suit ed to predict and fit kinetic trends with a minimal amount of data."
SionSwitzerlandEuropeAlkanesCybo rgsEmerging TechnologiesMachine LearningMethaneSwiss Federal Institute o f Technology Lausanne (EPFL)
NETL
NSTL
