| Titre : |
Using AI to rapidly develop new and improved high-performance coatings |
| Type de document : |
texte imprimé |
| Année de publication : |
2024 |
| Article en page(s) : |
p. 45-49 |
| Langues : |
Américain (ame) |
| Catégories : |
Analyse multivariée
Formulation (Génie chimique)
Intelligence artificielle
Modélisation prédictive
Recherche industrielle
Revêtements organiques
Simulation par ordinateur
|
| Index. décimale : |
667.9 Revêtements et enduits |
| Résumé : |
In the era of big data and digitalization, applying AI in product development is becoming essential for manufactures to stay competitive in their industry. Gone are the days where a limited group of domain experts plan and execute design of experiments (DOE) until a set of target properties are achieved. R&D teams across many industries are transforming their workflows by embracing AI initiatives to intelligently glean information from their historical, experimental, and product databases.
AI models built on past data can be used to reduce experimentation through simulation and achieve optimal properties much faster using numerical optimization. In this article, a high-performance coating dataset will be used along with ProSensus’ FormuSense software. The article will examine the typical steps to effectively utilize R&D datasets including :
1. Data preparation to ensure that the dataset is appropriate for model building. Common pitfalls such as missing data, insufficient variation, and data anomalies are investigated and resolved.
2. Model building using multivariate analysis and its intuitive plots that help subject matter experts interpret their dataset highlighting key correlations and trade-offs.
3. Simulation using the model to predict the outcome of potential experiments without running the actual physical experiment.
4. Constrained numerical optimization to calculate the ideal formulation and processing conditions required to achieve a targeted quality or set of qualities. |
| Note de contenu : |
- Predictive modeling : PLS for product development
- Constrained optimization
- Coatings case study : Assemble dataset - Mixture properties - Build & interpret PLS model - Simulate new experiments - Predicted quality - Model validity - Cost - Opitmmization
- Table 1 : Model validity metrics and formulation cost
- Table 2 : Ingredients used per class |
| En ligne : |
https://drive.google.com/file/d/12cX85i3IxmOIIQ0uBD58rEdYqkTmiOkV/view?usp=drive [...] |
| Format de la ressource électronique : |
Pdf |
| Permalink : |
https://e-campus.itech.fr/pmb/opac_css/index.php?lvl=notice_display&id=41093 |
in COATINGS TECH > Vol. 21, N° 3 (05-06/2024) . - p. 45-49
Aller sur edunet
Accueil
Using AI to rapidly develop new and improved high-performance coatings in COATINGS TECH, Vol. 21, N° 3 (05-06/2024)
