[article]
| Titre : |
Big data comes to coatings : Novel formulation optimisation using big data, modelling, and predictive tools |
| Type de document : |
texte imprimé |
| Auteurs : |
Partha Majumdar, Auteur ; Jonathan DeRocher, Auteur ; Michael Tran, Auteur ; Nipun Bisht, Auteur ; Rose Bohling, Auteur ; Adeline Ossola, Auteur ; Ivan Boronat Monfort, Auteur ; Philip Harsh, Auteur ; James Bohling, Auteur ; Jeff Sweeney, Auteur |
| Année de publication : |
2021 |
| Article en page(s) : |
p. 40-47 |
| Note générale : |
Bibliogr. |
| Langues : |
Anglais (eng) |
| Catégories : |
Agent mattant
Analyse des données
Données massives
Essais (technologie)
Formulation (Génie chimique)
Modélisation prédictive
Plan d'expérience
Revêtements organiques
|
| Index. décimale : |
667.9 Revêtements et enduits |
| Résumé : |
Big data analytics and machine learning (ML) have disrupted nearly every industry, and are now being embraced by the coatings sector, whose formulations offer an almost infinite formulation landscape. This paper reviews recent work in this area, explores the use of iterative experimental design to map the performance of a matting resin and discusses the use of advanced data analysis through the application of applying various algorithms. |
| Note de contenu : |
- EXPERIMENTAL : Materials
- FORMULATIONS
- TEST METHODS : Block - Tack - Coefficient of friction - Scrub - Stain
- OUTLINE OF DESIGN OF EXPERIMENTS (DOE)
- DATA ANALYSIS AND MODELS
- OPTIMISATION
- MODEL DEPLOYMENT
- Fig. 1 :Distribution of gloss, scrub, and scatter plot of gloss vs. scrub with CoF overlay
- Fig. 2 : Plots of LSMeans for comparing the impact of matting additives on (a) scrub and (b) CoF in white base, no extender formulations and the impact of both matting additives and extenders on (c) scrub and (d) CoF in white base with extender formulations
- Fig. 3 : Plots of LSMeans for comparing the impact of matting agents on (a) tack, (b) CoF, and (c) scrub in deep base, no extender formulations
- Fig. 4 : Plots of LSMeans for comparing the impact of interactions between matting agents and extenders on (a) tack, (D) CoF, and (c) scrub in deep base with extender formulations
- Fig. 5 : Expansion of initial input parameters from material type to characteristic properties
- Fig. 6 : Optimised formulation maps (a) all properties are equally important and (b) gloss, scrub, and CoF are twice as important as AKU and stain
- Fig. 7 : Interactive formulation map implemented in Plotly Dash and deployed in RStudio
- Table 1 : Typical white base formulations
- Table 2 : Typical deep base formulation
- Table 3 : Design variables including ingredient types and levels of interest
- Table 4 : Model R2 values of key paint properties
- Table 5 : Training and Test R2 values for the combined white base data set comparing methodologies applied to model paint properties using material property input parameters |
| En ligne : |
https://drive.google.com/file/d/1jepJbW7hRp6AOq_tJ4cjI3SSB2OMKBfh/view?usp=drive [...] |
| Format de la ressource électronique : |
Pdf |
| Permalink : |
https://e-campus.itech.fr/pmb/opac_css/index.php?lvl=notice_display&id=36482 |
in EUROPEAN COATINGS JOURNAL (ECJ) > N° 11 (11/2021) . - p. 40-47
[article]
|
Aller sur edunet
Accueil
Ajouter le résultat dans votre panier Affiner la rechercheAn eye on the process - an handle on quality in KUNSTSTOFFE INTERNATIONAL, Vol. 111, N° 5 (2021)
Industry 4.0 : Managing the digital transformation / Cham [Switzerland] : Springer International Publishing Switzerland (2018)
