terclim by ICS banner
IVES 9 IVES Conference Series 9 OIV 9 OIV 2024 9 Orals - Oenology, methods of analysis 9 Mineral wine profile: a major innovation in wine industry AI models

Mineral wine profile: a major innovation in wine industry AI models

Abstract

Multi-mineral wine profiling and artificial intelligence: implementing the signatures of each wine to train algorithms to meet the new challenges facing the wine industry.  Although their quantity is minimal, minerals are essential elements in the composition of every wine. Their presence is the result of complex interactions between factors such as soil, vines, climate, topography, and viticultural practices, all influenced by the terroir. Each stage of the winemaking process also contributes to shaping the unique mineral and taste profile of each wine, giving each cuvée its distinctive characteristics. This mineral composition, stable over time, is essential for an in-depth understanding of wines, and can serve as the basis for the development of new artificial intelligence algorithms.  In this context, we have created the mineral wine profile (mwp), which consists of a concentration profile of 41 mineral elements obtained simply by semi-quantitative analysis using inductively coupled mass spectrometry (icp-ms). Beyond the specific signature of a cuvée and the development of authentication services, the association of these signatures with other information (qualitative, soil, oenological, viticultural, environmental, etc.) Will make it possible, thanks to artificial intelligence, to create new solutions to the emerging challenges of the wine industry.  Minerals can be both the cause and/or consequence of the processes studied; their origins are complex, involving multiple chemical reactions at each stage, and their equilibria are relative between each element. Despite their great complexity and sensitivity, these profiles can be positioned at the heart of oenology guided by artificial intelligence. To date, we have analyzed and developed the world’s largest mineral database, currently containing over 20,000 mwps.  We will present the first proofs of concept illustrating the strength of this approach and discuss future perspectives.

Profilo multiminerale dei vini: un importante passo avanti nei modelli di IA dell’industria vinicola

Profilazione multiminerale dei vini e intelligenza artificiale: implementare le firme di ogni vino per addestrare algoritmi in grado di affrontare le nuove sfide dell’industria vinicola.  Sebbene la loro quantità sia minima, i minerali sono elementi essenziali nella composizione di ogni vino. La loro presenza è il risultato di complesse interazioni tra vari fattori come il suolo, i vitigni, il clima, la topografia e le pratiche viticole, tutti influenzati dal territorio. Ogni fase del processo di vinificazione contribuisce inoltre a plasmare il profilo minerale e gustativo unico di ogni vino, conferendo a ciascuna cuvée il suo carattere distintivo. Questa composizione minerale, stabile nel tempo, è di fondamentale importanza per una comprensione approfondita dei vini e può servire come base per lo sviluppo di nuovi algoritmi di intelligenza artificiale.  In quest’ottica, abbiamo intrapreso un’indagine sul mineral wine profile (mwp), che consiste in un profilo di concentrazione di 41 elementi minerali ottenuto semplicemente con un’analisi semi-quantitativa mediante spettrometria di massa accoppiata induttivamente (icp-ms). Oltre alla firma specifica di una cuvée e allo sviluppo di servizi di autenticazione, l’associazione di queste firme con altre informazioni (qualitative, pedologiche, enologiche, viticole, ambientali, ecc.) Consentirà, grazie all’intelligenza artificiale, di creare nuovi strumenti in grado di rispondere alle sfide emergenti dell’industria del vino.   I minerali possono essere sia causa che conseguenza dei processi studiati; le loro origini sono complesse, coinvolgono molteplici reazioni chimiche in ogni fase e i loro equilibri sono relativi tra ogni elemento. Nonostante la loro grande complessità e sensibilità, questi profili possono essere posizionati al centro dell’enologia guidata dall’intelligenza artificiale. Abbiamo già analizzato e sviluppato il più grande database di minerali, che attualmente contiene più di 20.000 mwp.  Presenteremo i primi proof of concept che illustrano la forza di questo approccio e discuteremo le prospettive future.

Profil multi-minéral des vins : avancée majeure dans les modèles d’IA viti-vinicoles

Profil multi-minéral des vins et intelligence artificielle : implémentation des signatures de chaque vin pour l’entraînement d’algorithmes au service des nouveaux enjeux de la filière viti-vinicole.  Bien que leur quantité soit minime, les minéraux sont des éléments essentiels de la composition de chaque vin. Leur présence découle d’interactions complexes entre divers facteurs tels que le sol, la vigne, le climat, la topographie et les pratiques viticoles, tous influencés par le terroir. Chaque étape du processus de vinification contribue également à façonner le profil minéral et gustatif unique de chaque vin, conférant ainsi à chaque cuvée son caractère distinctif. Cette composition minérale, stable dans le temps, revêt une importance fondamentale pour une compréhension approfondie des vins, pouvant servir de base au développement de nouveaux algorithmes d’intelligence artificielle.  Dans cette perspective, nous avons entrepris le recensement du mineral wine profile (mwp), qui consiste en un profil de concentration de 41 éléments minéraux obtenus simplement par analyse semi-quantitative en spectrométrie de masse à couplage inductif (icp-ms). Au-delà de la signature spécifique d’une cuvée et des développements de services d’authentification, l’association de ces signatures à d’autres informations (qualitatives, sols, œnologiques, viticoles, environnementales, etc.) Permettra, grâce à l’intelligence artificielle, de créer de nouveaux outils répondant aux défis émergents de la filière viti-vinicole.  Les minéraux peuvent être à la fois la cause et/ou la conséquence des processus étudiés ; leurs origines sont complexes, impliquant de multiples réactions chimiques à chaque étape, et leurs équilibres sont relatifs entre chaque élément. Malgré leur grande complexité et sensibilité, ces profils peuvent être positionnés au cœur de l’œnologie guidée par l’intelligence artificielle. Nous avons déjà analysé et développé la plus grande base de données minérales, contenant actuellement plus de 20 000 mwp.  Nous présenterons les premières preuves de concept illustrant la force de cette approche et discuterons des perspectives à venir.

Publication date: November 18, 2024

Issue: OIV 2024

Type: Article

Authors

Olivier Tillement¹, Coraline Duroux², Théodore Tillement², Leticia Gomes²

¹ Institut Lumière Matière, 10 Rue Ada Byron, Villeurbanne, France
² M&Wine, 305 Rue des Fours, Fontaines St Martin, France

Contact the author*

Tags

IVES Conference Series | OIV | OIV 2024

Citation

Related articles…

Reduce sulfur dioxide addition using a natural polymer chitosan phytate

Most oxidation reactions in wine require iron as a catalyst. The iron content of wine has decreased greatly in recent decades due to the use of low or no release cellar materials; however, in some cases it is still necessary to adopt winemaking practices to remove excess iron from wine, prevent its oxidation, and be able to reduce the addition of sulfur dioxide and other antioxidants.

Screening of phenolic compounds and antioxidant potential of grapes, wine and grape by-products

Polyphenols, bioactive secondary metabolites abundantly found in various grapevine components such as stalks, skins, and seeds, have attracted considerable attention in recent decades due to their potential health benefits. These compounds, including flavan-3-ols, flavanols, flavones, and stilbenes, are known for their antioxidant and anti-inflammatory properties.

Wine production as a source of positive impacts on the semi-natural environment

Wine is at risk of being labelled as a dangerous health product, based on studies that focus on its alcohol content. However, multiple studies suggest that moderate consumption is healthy. Changing the focus from health impacts to the environmental and socioeconomic impacts that wine companies cause, what can be said?

Chemical affinity and binding capacity between pre-purified Cabernet Sauvignon/merlot anthocyanins and salivary proteins monitored by UHPLC Q-TOF analysis

Apart from pro(antho)cyanidins and tannins, other phenolic compounds in wine or grapes have been shown to interact with salivary proteins and may contribute to overall sensory in-mouth sensations [1, 2]. Anthocyanins are the dominant phenolics in red wine and grape skin [3] , so it is expected that they come into contact and interact with salivary proteins after ingestion.

Correction of alcohol content in wines by partial vacuum evaporation during alcoholic fermentation

Climate change has become a reality that is becoming more and more apparent every day, with changes in the physico-chemical composition of grapes and an increase in the alcohol content of finished wines. These higher alcoholic degrees are not without consequences for the success of alcoholic and malolactic fermentation. Correcting the alcohol content (-20% of the initial alcoholic strength) is also part of an approach designed to meet consumer expectations for healthier, lighter or lower-alcohol wines (9 to 13% vol.). Correcting the alcohol content of wines also rebalances the mouthfeel by reducing the alcohol’s burn.