Recent progress on discovery and properties prediction of energy materials: Simple machine learning meets complex quantum chemistry

Kang, Youngnam; Li, Lejing; Li, Baohua

doi:10.1016/j.jechem.2020.05.044

Cited by 45 publications

(19 citation statements)

References 155 publications

(183 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Therefore, a variety of ML methods have been developed, including the Bayesian network, decision tree, artificial neural network (ANN), and support vector machines (Figure 2). [25] The fundamentals of these ML methods have been summarized well in many great reviews, [25, 28, 29, 31] and we hereby focus mainly on the applications of ML to battery research.…”

Section: An Overview On Machine Learning At Different Time and Length Scalesmentioning

confidence: 99%

“…Data can be generated much more easily from both experiments and theoretical calculations,whereas mining all the valuable information and hidden correlations behind the data by conventional approaches poses at remendous challenge.T herefore,avariety of ML methods have been developed, including the Bayesian network, decision tree,a rtificial neural network (ANN), and support vector machines (Figure 2). [25] Thef undamentals of these ML methods have been summarized well in many great reviews, [25,28,29,31] and we hereby focus mainly on the applications of ML to battery research.…”

Section: An Overview On Machine Learning At Different Time and Length Scalesmentioning

confidence: 99%

“…Thes tructure-function relationship of ac omplicated system can, therefore,b ed iscovered and unveiled more efficiently.More importantly,anovel research approach has been established based on ML methods.T his statistically driven design is completely different from conventional theoretical approaches,which mainly involve structure-property calculations or crystal structure prediction. [23] Then umerous datasets generated through high-throughput calculations and experiments are fed into ML to discover valuable information and hidden correlations,w hich can be quite challenging in current physical science.B ased on highquality datasets,the ML models have been shown to have the ability to predict the physicochemical properties of materials (such as ionic conductivity and viscosity [24] )a nd assist the design of functional materials (such as drugs, [25,26] energy materials, [27,28] porous materials, [29] and small molecules [30] ). Consequently,experimental, theoretical, and data tools have become three indispensable methods in current scientific research and are displaying great potential in battery studies (Figure 1).…”

Section: Introductionmentioning

confidence: 99%

“…The numerous datasets generated through high‐throughput calculations and experiments are fed into ML to discover valuable information and hidden correlations, which can be quite challenging in current physical science. Based on high‐quality datasets, the ML models have been shown to have the ability to predict the physicochemical properties of materials (such as ionic conductivity and viscosity [24] ) and assist the design of functional materials (such as drugs, [25, 26] energy materials, [27, 28] porous materials, [29] and small molecules [30] ). Consequently, experimental, theoretical, and data tools have become three indispensable methods in current scientific research and are displaying great potential in battery studies (Figure 1).…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Applying Machine Learning to Rechargeable Batteries: From the Microscale to the Macroscale

Chen

Shen

et al. 2021

Angewandte Chemie

View full text Add to dashboard Cite

Emerging machine learning (ML) methods are widely applied in chemistry and materials science studies and have led to a focus on data‐driven research. This Minireview summarizes the application of ML to rechargeable batteries, from the microscale to the macroscale. Specifically, ML offers a strategy to explore new functionals for density functional theory calculations and new potentials for molecular dynamics simulations, which are expected to significantly enhance the challenging descriptions of interfaces and amorphous structures. ML also possesses a great potential to mine and unveil valuable information from both experimental and theoretical datasets. A quantitative “structure–function” correlation can thus be established, which can be used to predict the ionic conductivity of solids as well as the battery lifespan. ML also exhibits great advantages in strategy optimization, such as fast‐charge procedures. The future combination of multiscale simulations, experiments, and ML is also discussed and the role of humans in data‐driven research is highlighted.

show abstract

Section: An Overview On Machine Learning At Different Time and Length Scalesmentioning

confidence: 99%

Section: An Overview On Machine Learning At Different Time and Length Scalesmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Applying Machine Learning to Rechargeable Batteries: From the Microscale to the Macroscale

Chen

Shen

et al. 2021

Angewandte Chemie

View full text Add to dashboard Cite

show abstract

“…There is currently some movement in the field, and DoE has recently been successfully combined with machine learning [252]. DoE-inspired machine learning has also been used alone [253][254][255]. I believe these types of studies will be more plentiful in the not so far future, as especially artificial intelligence is a hot research topic.…”

Section: Reflectionsmentioning

confidence: 99%

Area-selective atomic layer deposition of molybdenum oxide

Kvalvik

Borgersen

Hansen

et al. 2020

Journal of Vacuum Science &Amp; Technology A: Vacuum, Surfaces, and Films

View full text Add to dashboard Cite

have supervised the work.Per-Anders and Ola, there is no way this thesis could be finished without you. Per-Anders, you are truly an excited person and I have never met anyone so passionate about his or her own field as you. That passion is contagious, and so is your cheerfulness and joy to share your knowledge. Ola, you are the most understanding and supporting person on the planet. Your response time is amazingly fast and you take all stupid and not so stupid questions seriously. You see every find as an opportunity, and that is great. When I saw an ALD system not behaving, only depositing on certain surfaces, you saw the chance to explore an area-selective system. I am so deeply grateful to both of you.I am also very grateful to Øystein Evandt and his insight on Design of Experiments (DoE), that he is always more than willing to share. Moreover, a special thank you goes to Dr. David Wragg for his aid with Pawley refinements and to my co-author on Paper I, Jon Borgersen. Making the world's second smallest university logo would not have been possible without you. Augustinas Galeckas, co-author on Paper III, has my sincere gratitude for a swift response and aiding with various optical measurements.Thank you all.During my many years at NAFUMA I have had a lot of nice colleagues and even shared office with many of you. I have probably been of the more talkative kind of office mate, but I think most of you managed ok. Thanks for making work-life good. Special thanks go to Kristian x2, Leva, Michael and Katja, for particularly good company.The largest thank you goes to the people surrounding me. Luckily, that is a lot of people and I am so blessed to have each and every one of you. I am looking forward to times where we can all gather up for large family dinners again, although fitting 25+ people in our house is kind-of tight. Special thanks go to my parents for their endless support and babysitting. You have thought me persistence, perfectionism and hard work, and always taken for granted that I would be able to reach my goals.Thanks for having that belief, also at times where I lost it myself. A huge thank you goes to my in-laws for cheering and babysitting, including my always calm, positive and caring sister-in-law. I also want to thank the engineers in my family, particularly my grandfather, for inspiring me towards a scientific career. Kaja also deserves a huge thank you, simply for existing. Thanks for being my best friend since we started schooland staying with me until now when I finally finish my formal education.Many times during this process, I have been told that pursuing a PhD whilst having three kids, is something you just don't do. It pisses of my inner feminist, but they do have a point, combining sleepless iv nights, dreadful pregnancies and having to make it home before the kindergarten closes whilst pursuing a phd, is not always a primrose path. My favorite counterargument for this is that children give you the best possible mental timeouts from your work. And trust me, in this process, you need m...

show abstract

Machine Learning‐Based Assessment and Optimization of Electrode Materials for Supercapacitors

Moharana,

Sahu,

Meher

et al. 2024

Nanostructured Materials for Energy Storage

View full text Add to dashboard Cite

Recent progress on discovery and properties prediction of energy materials: Simple machine learning meets complex quantum chemistry

Cited by 45 publications

References 155 publications

Applying Machine Learning to Rechargeable Batteries: From the Microscale to the Macroscale

Applying Machine Learning to Rechargeable Batteries: From the Microscale to the Macroscale

Area-selective atomic layer deposition of molybdenum oxide

Machine Learning‐Based Assessment and Optimization of Electrode Materials for Supercapacitors

Contact Info

Product

Resources

About