Method and device for improving the quality and traceability of alcoholic beverages, in particular wines
Patent Information
- Application Number
- EP2022854161
- Authority / Receiving Office
- EP · EP
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2021-12-29
- Filing Date
- 2022-12-27
- Publication Date
- 2024-11-06
AI Technical Summary
Current methods for improving the quality and traceability of wines are insufficient, as they often rely on limited analytical techniques that require complex and specific analyses, failing to effectively manage and monitor the production, quality, organoleptic properties, authenticity, and traceability of wines, especially in detecting counterfeits and variations in metal concentrations.
A high-performance process implemented by computer, using statistical processing and artificial intelligence to analyze the mineral composition of wines, particularly metals, by collecting and storing samples in inert containers, and processing data to monitor and manage various stages of wine production, from agricultural production to commercial distribution, without requiring additional products or complex analyses.
This approach provides a reliable marker for the mineral profile of wines, enabling comprehensive monitoring and optimization of wine production, quality control, authenticity verification, and traceability, enhancing oenological practices and reducing counterfeiting, while maintaining the stability of metal compositions over time.
Smart Images

Figure 1.1
Abstract
Description
DESCRIPTION Title: Process and device for improving the quality and traceability of alcoholic beverages, in particular wines Field of the invention
[0001] The invention relates to the field of the production of alcoholic beverages, in particular wines, from a raw material derived from agriculture, in this case from viticulture in the case of wines.
[0002] The invention relates more specifically to a technology which aims to improve the quality and traceability of alcoholic beverages, in particular wines (the term "wine" hereinafter designating any alcoholic beverage), by automatic or semi-automatic methods based on statistical processing (of the datamining type), possibly carried out using artificial intelligence tools, of analytical data relating to the mineral, in particular metallic, composition of these beverages (wines).
[0003] To this end, this technology includes a process, in particular implemented by computer, for managing and / or monitoring: * the agricultural production of a raw material for the production of an alcoholic beverage, preferably wine; * the manufacture of this alcoholic beverage; * the quality of the alcoholic beverage; * the organoleptic properties of the alcoholic beverage; * the authenticity of the alcoholic beverage; and or * the traceability of the alcoholic beverage.
[0004] The invention also relates to a device for implementing this method. Technological background of the invention
[0005] Alcoholic beverages and in particular wines have a significant economic impact. In 2020, the global alcoholic beverages market corresponds to a turnover of 1.470 billion dollars. At the top are beers, followed by spirits, wines and finally ciders. In the European Union, the combined annual sales of spirits and wines represent approximately 38 billion euros, in 2020.
[0006] In this area, consumer expectations relate to the quality of alcoholic beverages, health safety, access to information on production conditions and guarantees in terms of origin, identity and authenticity of these drinks.
[0007] Professionals in this sector are constantly seeking to improve product quality, without sacrificing tradition. In this context, the aim is to limit the risk of defects and to offer new sensations adapted, as closely as possible, to consumer tastes.
[0008] To do this, it is useful to have various and varied indicators, which make it possible to monitor and control the entire sector, from the production of the agricultural raw material, through the manufacture of the alcoholic beverage, to its storage and maturation in various containers.
[0009] Reference is made here to traceability, which is the ability to trace the path of a foodstuff throughout its production and distribution chain, either from the first origin of the product to its delivery to the final consumer, or as so goes the saying "from the fork to the fork".
[0010] Traceability is also a requirement of the standards relating to the ISO 9000 & 9001 quality management system.
[0011] Traceability also makes sense with regard to consumer health and safety. It provides them with reliable information on the substances present in alcoholic beverages. It also guarantees consumers perfect consumer safety, from purchase to the end of the product's life.
[0012] Traceability is also crucial when it comes to the origin, provenance and logistics of alcoholic beverages. It gives access to information on the production region of the agricultural raw material of the alcoholic beverage, for example wine, as well as on the places and conditions of manufacture, storage and transport of the alcoholic beverage.
[0013] With regard to the identity and authenticity of these alcoholic beverages, it is necessary to compare the figure of 38 billion euros in annual sales of wines & spirits in the European Union, in 2020, 1.3 billion euros (or 3.3%) of losses due to the presence of counterfeit products on the market. These missed sales translate into the direct loss of approximately 4,800 jobs in Europe.
[0014] Beyond the legal means of combating this scourge, players in the alcoholic beverages sector have implemented physical means of authenticating bottles. Examples include RFID chips, indelible marking, electronic chips, holograms, tamper-proof seals and barcodes.
[0015] Traceability for the detection of counterfeits, falsifications or contaminations also involves the characterization in analytical chemistry of the alcoholic beverage, taken as such.
[0016] Alcoholic beverages, especially wines, contain mineral elements, including metals and metalloids. Wines usually contain (i) major elements such as Ca, K, Mg, and Na -10-1000 mg / L-, (ii) minor elements: Al, Fe, Cu, Mn, Rb, Sr and Zn, -0.1-10 mg / L- and (iii) traces -i.a- Ba, Cd, Co, Cr, Li, Ni, Rb, and V -0.1-1000 μg / L.
[0017] Their presence has multiple origins: cultivation, vinification and breeding. These elements influence the quality and the final characteristics of the wine, at an oenological level: taste, aromas, evolution, quality, feeling, oxidation... and can have a positive (or negative) impact on the feeling of consumers and the price of final sale.
[0018] Neglected for a long time, or relayed to simple regulatory positions vis-à-vis health standards to be respected (maximum concentrations to be respected), in favor of the multiple organic compounds of wine, metals are nevertheless at the center of the life of wine.
[0019] The use of the concentration of metals associated with predictive statistical analyzes and / or other artificial intelligence tools has already been proposed to try to identify or trace families of wines.
[0020] Nevertheless, these tools are often insufficiently efficient, they have been developed on weak samples (a maximum of a few hundred different wines), limited to a region and neglect many parameters. The results, even if they have made it possible to identify trends, often remain limited to specific regions and require very specific analyzes at the scale of ultra-traces and / or isotopic ratios of certain elements. Their performances are still insufficient and these approaches remain very little used, they often require very complex analyzes to be reliable. They cannot be used in comparison, nor to establish wine categories or classifications, nor to exclude a wine from well-defined wine categories. Objectives of the invention
[0021] In this context, the invention aims to satisfy at least one of the following objectives:
[0022] - provide a high-performance process, in particular implemented by computer, for managing and / or monitoring the production of agricultural raw materials useful for the manufacture of alcoholic beverages, the manufacture of these alcoholic beverages, the quality of these alcoholic beverages , the organoleptic properties of these alcoholic beverages, the authenticity of these alcoholic beverages, and / or the traceability of these alcoholic beverages.
[0023] - provide a high-performance process, in particular implemented by computer, for managing and / or monitoring wine production, wine production, wine quality, wine organoleptic properties, wine authenticity, and / or wine traceability.
[0024] - provide a high-performance and economical process, in particular implemented by computer, for managing and / or monitoring the production of agricultural raw materials useful for the manufacture of alcoholic beverages, the manufacture of these alcoholic beverages, the quality of these alcoholic beverages, the organoleptic properties of these alcoholic beverages, the authenticity of these alcoholic beverages, and / or the traceability of these alcoholic beverages.
[0025] - provide a high-performance and economical process, in particular implemented by computer, for managing and / or monitoring wine production, wine production, wine quality, wine organoleptic properties, wine authenticity , and / or wine traceability.
[0026] - provide a high-performance, economical and reliable process, in particular implemented by computer, for managing and / or monitoring the production of agricultural raw materials useful for the manufacture of alcoholic beverages, the manufacture of these alcoholic beverages, the quality of these alcoholic beverages, the organoleptic properties of these alcoholic beverages, the authenticity of these alcoholic beverages, and / or the traceability of these alcoholic beverages.
[0027] - providing a high-performance, economical and reliable process, in particular implemented by computer, for managing and / or monitoring wine production, wine production, wine quality, wine organoleptic properties, authenticity of wines, and / or traceability of wines.
[0028] - provide a high-performance, economical and reliable device for the implementation of the process referred to in the above objectives. Description of the invention
[0029] DEFINITIONS
[0030] Throughout this presentation, any singular designates either a singular or a plural.
[0031] The definitions given below as examples may be used to interpret this presentation: - "alcoholic drink": drink prepared by alcoholic fermentation of a vegetable raw material, preferably agricultural. It may in particular be a wine, a beer, a cider, a liqueur, a spirit, a whisky, a brandy, a tequila, a vodka, a rum, a cognac, an armagnac, an Asian alcohol, etc. - "inert container with respect to the alcoholic beverage"; container which retains its physical integrity and good mechanical properties (breaking limit and plasticity still sufficient for reliable use as a container for liquids) after 10 years of contact with the alcoholic beverage and does not react with the latter, in particular the pollutant in particular with metallic elements - "quality of the alcoholic beverage"; this covers in particular the organoleptic properties, such as the taste, the smell, the structure, the texture, the balance, the color, the aspect, the consistency, the length in the mouth..., but also the sanitary qualities independently ethanol-related toxicity, such as the content of heavy metals such as lead or cadmium; -"significantly"; "of the order of" ; "approximately": to within + or - 10%, preferably within 5%.
[0032] PROCESS The invention satisfies at least one of the above objectives and relates, according to a first aspect, to a method, in particular implemented by computer, for managing and / or monitoring at least one factor *f x * chosen from a set of factors including: *f 1 * the agricultural production of a raw material for the manufacture of an alcoholic beverage, preferably wine, "f 2 * the manufacture of this alcoholic beverage; "f 3 * the conservation of this alcoholic beverage; "f 4 * the maturation of this alcoholic beverage; *f 5 * the consumption of this alcoholic beverage *f 6 * the quality of this alcoholic beverage; *f 7* the authenticity of this alcoholic beverage in relation to a reference chosen from the group comprising -advantageously constituted by-: the names of the wine and of the estates, the designations of origin; geographical indications; traditional specialties guaranteed; labels; registered trademarks; and their combinations; *f 8 * the traceability of this alcoholic beverage; *f 9 * the selling price of this alcoholic beverage; this process consisting essentially of: (a) collect at least one, preferably at least two samples of the alcoholic beverage, place each of them in an inert container with respect to the alcoholic beverage and close the latter tightly with an equally inert stopper; (b) assigning to each sample data on the alcoholic beverage, chosen from the group comprising - advantageously consisting of - data on the origin, data on the terroir, data on the manufacture, data on the conservation and the maturation, consumption data, physico-chemical data, qualitative data, in particular organoleptic data, economic data, commercial data... and combinations of these data; (c) storing at least some of the samples collected in step (a), under determined conditions; (d) analyzing each sample to determine at least one mineral, preferably metallic, profile; (e) constituting a database relating to the samples and resulting from step (b) and from step (d); (f) optionally, complete and / or update the data assigned in step (b), at least once, on all or part of the samples; (g) optionally, completing and / or repeating the analyzes carried out in step (d), at least once, on all or part of the samples; (h) process these data by statistical analysis, advantageously by automatic or semi-automatic methods based on statistical processing, and, even more advantageously by data mining or "datamining", preferably using tools for artificial intelligence and / or other "datamining" techniques; (i) and use the processed data for the management and / or monitoring of at least one of the aforementioned *f** factors.
[0033] It is to the merit of the inventors to have proposed, among other things, to take, reference, and keep samples of alcoholic beverages, for example wine, and to determine a mineral composition (e.g. chosen from metals, metalloids, halogens , P, S, Se) reference. Indeed, this principle according to the invention of storing samples of alcoholic beverages, over a long period of time (several years), to have a reference is counter-intuitive. It is known in this field that alcoholic beverages are the site of chemical and often even biochemical reactions which give them an evolutionary character. This evolution can lead to a maturation in the direction of an improvement of the quality, but also to degradations which affect the organoleptic properties, even the food safety of the alcoholic beverages considered. The inventors have nevertheless bet on the fact that the mineral composition of an alcoholic beverage benefits from stability over time, unlike the constituent organic compounds of this alcoholic beverage.
[0034] To do this, the inventors have judiciously used inert containers, non-contaminating and forming a barrier, with respect to the mineral profile of the alcoholic beverage, and moreover, to containers each containing a sample of alcoholic beverages and capable of being closed tightly with a plug. This makes it possible to maintain over time the elementary mineral composition of the samples and / or at the very least, elementary mineral composition ratios.
[0035] This new and inventive approach gives access to a mineral profile, preferably metallic, of the alcoholic beverage, e.g. wine, which constitutes a reliable marker allowing the control, the management and / or the monitoring of a plurality of stages of the complete chain, from agricultural production to commercial distribution, of the alcoholic beverage, as well as a certain number of states of the latter, before its consumption. This reliable marker can be made up of several dozen different chemical elements, for example more than 50 different chemical elements which provide a considerable wealth of information.
[0036] In the wine sector, the stages of the supply chain include growing the vines, harvesting and all the winemaking processes, maturing, aging, packaging (bottling) and storage. The mineral (metallic) profile of wines is also a reliable reflection of the characteristics, sanitary, regulatory, origin and authenticity, as well as the quality and excellence of the wines.
[0037] The mineral (metallic) profile of the alcoholic drink -wine- is all the more interesting according to the invention, that it is refined / optimized by statistical analysis, even preferably by artificial intelligence algorithms, and that it is used as a tool for classifying and predicting properties of alcoholic beverage, especially vinous properties when it comes to wine.
[0038] This fine characterization gives access to a whole range of improved statistical or predictive models for alcoholic beverages.
[0039] On the strength of this, a whole panoply of possibilities for corrective action opens up for the producer of the alcoholic beverage in terms of the agricultural production of the raw material, the harvesting of this material, the manufacture of the alcoholic beverage. from this raw material, the conservation and consumption of this alcoholic beverage. This leads to notable oenological optimizations in the wine sector. These new fields of action are very promising for this economic sector.
[0040] The method according to the invention also represents substantial progress in the quest for traceability, which makes it a formidable anti-counterfeiting weapon and an unequaled means of monitoring the quality of alcoholic beverages, in particular wines. It differs from the existing one in that it does not require any specific addition of products within the drink and / or its containers and / or packaging and / or labeling. Finally, it gives the producer of the alcoholic beverage better control of all the parameters of the sector, from agricultural production, terroir, manufacturing, maturation, conservation, to consumption. In other words, it gives the keys to strive for and achieve excellence. The alcoholic beverages, in particular wines, obtained by implementing this process, undoubtedly have an increased commercial value.
[0041] Alcoholic beverages are characterized by a mineral signature or profile, in particular metallic. This profile originates, in particular, in the plant raw material, in its culture medium and / or in its cultivation and in all or part of the stages in the manufacture of the alcoholic beverage, namely in the case of wine: vinification, aging , aging, maturation, conservation, until consumption. This evolution of the mineral profile, in particular metallic, results in the appearance and disappearance, through variations in concentration, of the metallic elements of the alcoholic beverage. In a first part, the method according to the invention is based: * on the regular observation of this tracer, which is the mineral (metallic) profile, throughout its evolution; * on the forecast of this evolution, * as well as on the implementation of means of intervention to control this evolution and bring the alcoholic drink towards the desired quality. In a second part, the method according to the invention is intended to use the profile of the metallic signature of the alcoholic beverage for what it is, namely a tracer of the health & food state and a marker of the identity alcoholic drink.
[0042] Step (a): Collection
[0043] In this step, it is a question of taking at least one, or even at least 2 samples, of at least 1 ml, for example 50 ml, of the alcoholic beverage and placing each sample in an inert container with respect to the alcoholic beverage. , then to close this container hermetically with a stopper, also inert vis-à-vis the alcoholic beverage.
[0044] Step (a): In a preferred embodiment of the method according to the invention, the number N of alcoholic beverages collected in step (a) is such that, in increasing order of preference: N≥500; N S 1,000; N ≥ 10,000. Preferably, the volume -Ve- of alcoholic beverage taken for each sample is such that, in increasing order of preference with Ve expressed in milliliters: 0.5≤Ve≦100; 1 ≤ Ve≤ 50; 10 ≤ Ve≤ 30.
[0045] Step (a): It is appropriate, according to the invention, for the inert container to be distinguished by a content -Tm- for each of the following mineral elements: Ca; mg; Zn; Fe; Mn; Cu; Al; Whether ; Neither ; V; N / A ; P; Co; Cr; K; Li; Pb; Se; CDs; Kg; As; , such that, in increasing order of preference with Tm measured according to a measurement method M tm and expressed in ppb: Tm≤5; Tm≤ 3; Tm≤ 2; Tm≤ 1 The measurement method is described below: the container is filled to more than 50%, and preferably between 60 and 75%, of its capacity, with 1% nitric acid, and maintained at 50°C for 12 hours, the solution obtained is then taken for elemental analysis. The analysis is carried out using an ICP-MS Agilent® 7700. This is equipped with an octopole which will be used in helium collision mode (He flow rate = 4.3 mL / min) before entering the quadrupole type mass analyzer. Each sample is then nebulized using a Micromist® nebulizer then introduced into the ICP-MS after passing through a Scott chamber, cooled to 2°C. An equilibration and signal stabilization time of 60 s is programmed before proceeding to the actual measurement.
[0046] Step (a): In accordance with a remarkable feature of the invention, the container and its stopper are made of a material chosen from thermoplastic polymers, preferably polyolefins, and more preferably still from the group comprising -preferably consisting of- polyethylene, polypropylene and mixtures thereof, with polyethylene being a preferred cap material and polypropylene being a preferred container material. Alternatively, the container could be made of silica.
[0047] According to an advantageous embodiment of the invention, the samples are taken at different stages of development of the alcoholic beverage. For example, in the wine sector, samples are taken: during storage in vats, before bottling and / or after bottling and / or at different times after bottling (several months or several years after bottling).
[0048] Advantageously, the container preserves the purity of the sample, in particular its inorganic purity, and also efficiently gathers the data related to this sample. To guarantee good readability of this information over time, barcode / QR Code labeling is an interesting solution. The data of the sample are in particular: date, sampler, technician in charge of the analysis, results... They can be read, for example, from a simple "scan" in all conditions.
[0049] According to a remarkable method of step (a), the taking of a sample from a container, such as a tank, a barrel or a bottle, consists in emptying at least 10%, preferably at least 1% and at least plus 50% of the volume of alcoholic beverage present in the sampling container, before taking the sample.
[0050] Step (b): The data assigned to the samples in this step is preferably: * origin data includes name of alcoholic beverage, name of producer, name of estate, year of production, date of sampling, name of cuvée, batch number, and / or type of alcoholic beverage; * the data on the terroir includes the designation of origin, the geographical indication, the country, the region, the place, the plot, the grape varieties, the distribution of the grape varieties, the exposure, the sunshine, the density of planting (in vines / ha), the type of vine pruning, the cultivation method, the fertilization of the vine, the grass cover, the pest control, the watering, the average age of the vine, the relief, soil type, water source, and / or vine irrigation; * the data on production includes the type of harvest, the date of the harvest, the type of sorting and destemming, the type of vinification, the type of press, the maceration time of the skins and pips in the must, the material tanks, the addition of yeast or not, the type of fining and clarification, the filtration system, and / or the assembly; * storage data includes storage time, successive container types, capacity, temperature, humidity, closure type, and / or packaging date; * consumption data includes the presence and level of sulphites, the presence and level of phenolic compounds, the percentage of alcohol, and / or the presence and level of aromatic compounds; * qualitative data, in particular organoleptic, include evaluations of balance, length, intensity, complexity (and / or concentration) and typicity [ELIC(T) method]; * the organoleptic data includes the color, the aromas, the tastes, and / or the duration of expression in the mouth of the aromas of the wine, preferably expressed in caudalies, * economic data includes price, sales volume, and / or sales amount; * commercial data includes labels, competition awards / medals, rankings, and / or scores received. These different types of data can be classified in files which will be organized in the database referred to in step (e) of the method according to the invention.
[0051] It is advantageous that access to this data on the sample is easy and fast over the entire sample life.
[0052] Step (c): This step (c) involves collecting several samples of alcoholic beverages, in particular different wines (more than 100, or even more than 1,000, or even more than 10,000) in a cupboard or storage room to establish a data bank. reference samples, usable for several years (for example at least 5 years, or even at least 10 years). A sample library of alcoholic beverages, in particular wines, is thus constituted, in which each sample retains its elemental mineral, in particular metallic, content. This stability opens the door to applications for health safety (dosage of heavy metals, toxicity), for the fight against fraud or counterfeiting or for subsequent additional analyses, in connection with the quality of a wine. Indeed, all the analyzes are not necessarily carried out soon after the sampling, for reasons of economy.
[0053] Advantageously, the storage conditions for the samples collected in step (a) are as follows: temperature ≤40° C.; pressure≤ 1 to 2 bar; humidity £90%; duration at one month, preferably ≥ one year, and, even more preferably at 5 years.
[0054] Step (d): It is preferable for this step to include the analysis of at least one reference metallic element, within a period less than or equal to 3 months, preferably less than or equal to 1 month, after sampling.
[0055] Step (d): The mineral profile analyzed in this step for the samples preferably comprises: * at least 5 mineral elements (called main) chosen from B, Na, Mg, P, S, Cl, K, Ca, * at least the following metallic elements (called trace metals): Fe; Cu; Zn; Min; * optionally at least one of the following metallic elements: Pb and Cd; * at least 10, preferably at least 30, and, even more preferably, at least 40 elements, for example between 50 and 100, chosen from the following trace mineral elements: Rb, Cs, Sr, Ba, Ce, Ti, V , Cr, Co, Ni, Zr, Mo, Ag, Al, Ga, Sn, As, Br, I, Se; and / or from the following ultra-trace mineral elements: La, Pr, Nd, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb, Lu, Th, U, Sc, Y, Nb, Ru, Rh, Pd, Hf, Ta, W, Re, Os, Ir, Pt, Au, Kg, Tl, Bi, Sb; * possibly of at least part of the isotopes of these elements * measurements of the concentrations of these elements; * and / or the concentration ratios of these elements, and possibly of all or part of their isotopes. It should be noted that iron, copper, zinc and manganese are the main trace elements that have an impact on the life of the wine. According to one possibility of the invention, the analytical data can be collected on an analytical sheet specific to each alcoholic beverage (wine).
[0056] To optimize the method according to the invention, it is preferable that the elements chosen for the analysis be the same for all the drinks analyzed.
[0057] Step (d): In this step, in addition to the absolute content of mineral, preferably metallic, elements, we are interested in the relative contents between mineral chemical elements, preferably metallic. This relative analytical variant is particularly appropriate for countering the effects of dilution or evaporation. It also allows better sensitivity in the detection of stable areas of metal concentrations in wine, areas which would be less influenced by bottle contamination and / or by precipitation which could lead to selective modifications of some metal concentrations over time.
[0058] Step (d): In a preferred mode of implementation of the invention, where the alcoholic beverage is a wine, step (d) advantageously comprises the analysis of chemical and physical parameters of the alcoholic beverage, these parameters being , preferably chosen from the group comprising - advantageously consisting of -: TAV (Alcoholic Strength by Volume), glucose + fructose, AT (Total Acidity), acetic acid, SO 2 free, SO 2 total, pH, SO 2 active, ethanal, malic acid, lactic acid, CO 2 , acid manufactures, gluconic acid, glycerol, optical density (absorbance at one or more wavelengths 280, 420, 520 and 620 nm), and all combinations of these parameters. These chemical and physical parameters may be subject, in the same way as the analytical data comprising at least one mineral profile of the alcoholic beverage, to processing by statistical analyzes in accordance with step (h), possibly assisted by a system of artificial intelligence.
[0059] It is particularly advantageous in the context of the invention to verify the relevance of the analyzes of all or part of the samples concerned. To do this, it is possible to carry out comparative analyzes of the concentration, of at least one given target mineral element, between the undiluted sample and at least one dilution of this sample and / or between at least 2 dilutions of this sample. The solvent used for the dilution is advantageously an aqueous solvent, e.g. chosen from the group comprising aqueous acid solutions, such as nitric acid. The dilutions used can be: a 1v / Xv dilution and a 1v / 2X dilution, with positive natural integer X between 1 and 10. For example, we can have: 1v / 5v dilution and a 1v / 10v and / or 1v / 10v dilution and a 1v / 20v dilution. In accordance with the invention, it is considered that the analysis of the concentration of the mineral element considered is reliable if the ratio between the concentration c 1v / Xv measured at dilution 1 v / Xv on concentration C 1v / 2xv mesur ed at dilution1 v / 2xv , is such that: 1 ≤ C 1v / Xv / VS 1v / 2Xv 3; preferably 1.5 C 1v / Xv / VS 1v / 2Xv 2.5.
[0060] Step (e): Advantageously, the database constituted during step (e) groups together data relating to at least 500, preferably at least 1000, and, more preferably still, to at least 10,000 different alcoholic beverages. For each alcoholic beverage, for each wine, the database gathers the analytical data from stage (d) and the data attributed to stage (b) on the origin, on the terroir, on the production, on the conservation, consumption, quality, organoleptic properties, economic aspects and commercial aspects. This database advantageously operates dynamically, thanks to regular updates. [00611 Step (h): Advantageously, the processing of the data according to step (h) essentially consists in → use at least one of the following means: * exploratory analyzes and logistic regressions to perform classifications based on Principal Component Analysis (PCA or PCA) * discriminant analyzes (ADL or LDA in English for Linear Discriminant Analysis) * predictive model algorithm, preferably chosen from the group comprising -ideally constituted by-: Random forest (RF) Decision Tree Forest and / or Artificial neural networks (ANN) and / or Support - Support vector machines (SVM); to perform at least one of the following actions: * Identify and predict the origins of the different mineral elements within a wine; * Identify and predict the impacts on the quality of a wine and the evolution of the quality of a wine of the different mineral elements within a wine; * Adjust the origins to adapt the quality, by choosing, for example, grapes from plots with the best probability of a metallic profile corresponding to the desired quality
[0062] Discriminant analyses:
[0063] Exploratory analyzes and logistic regressions for classifications are based on Principal Component Analysis (PCA) or, depending on the field of application, the Karhunen-Loève transformation (KLT)1. is a method that consists of transforming variables linked together (called "correlated" in statistics) into new variables uncorrelated from each other. These new variables are called "principal components", or principal axes. It allows the statistician to reduce the number of variables and make the information less redundant.
[0064] This is an approach that is both geometric (the variables being represented in a new space, along directions of maximum inertia) and statistical (the search for independent axes best explaining the variability — the variance — of the data).
[0065] Linear Discriminant Analysis (LDA) is one of the predictive discriminant analysis techniques. It is a question of explaining and predicting the membership of an individual in a predefined class (group) from its characteristics measured using predictive variables.
[0066] Linear discriminant analysis can be compared to supervised methods developed in machine learning and logistic regression developed in statistics.).
[0067] Such analyzes can be performed using toolboxes or libraries in many software environments such as Python (Sdkit Leam library), Matlab (Statistics and Machine Learning Toolbox toolbox), R (statistics package for PCA, MASS package for LDA and net package for logistic regression).
[0068] The analysis techniques used here are simple, they are based on a basis that allows easy understanding, allow an interpretation of the relationship between the selected elements since the techniques are based on a linear analysis and they are widely available in most statistical software packages.
[0069] Predictive model algorithm:
[0070] The Random Forest (RF) decision tree analysis technique has been widely used in many scientific fields in recent years (Ga'al et al., 2012). This statistical learning theory was proposed by Breiman in 2001 (Breiman, 2001; Tian et al., 2017). Random forests consist of a group of predictor trees where each tree describes a subset of data, which have been sampled according to different observations and according to different variables (Breiman, 2001). The final prediction obtained by the classification forest is the majority vote obtained by consulting for all the decision trees (Tian et al., 2017; Zahiri et al., 2013). When random forest operates in regression mode, the prediction is the average of all predicted values (Palmer et al., 2007; Vigneau et al., 2018). These types of models can be used for a wide variety of applications, for example to classify invasive plant species (Cutler et al., 2007), to detect Medicare fraud (Bauder and Khoshgoftaar, 2018) or to predict disease progression. idiopathic pulmonary fibrosis using CT scans (Shi et al., 2019).
[0071] Artificial neural networks (ANN):
[0072] Artificial neural networks are automatic learning methods that can improve their behavior with new observations, i.e. with experience (Anjos et al., 2015). ANNs are made up of an interconnected group of nodes (called artificial neurons) that process information (Anjos et al., 2015). A neuron in the network operates according to a simple rule that combines its inputs into outputs. For example, a neuron can calculate the sum of its input signals and responds with an output signal by comparing whether this sum is greater than or equal to a threshold value. In the network, the organization of connections makes it possible to define different types of networks: proactive, recurrent networks, etc. Finally, transmission of signals from one neuron to another can be modulated by learning rules (in analogy with the neuronal plasticity of the central nervous system). In particular, neurons can be connected by layers, an input layer (which receives the input data), one or more intermediate layers and a final layer which generates the output (variable) signals (Anjos et al., 2015). We can also work with so-called 'hidden' neurons (Linares-Rodriguez et al., 2013). They are able to extract meaningful features from data and they can 'learn' the relationship between inputs and outputs when there is enough training data (in quantity and complexity) (Chiang and Chang, 2009) . Again, these models can be used for a wide variety of applications, for example discriminating the botanical origin of different honey samples (Anjos et al., 2015), modeling rainfall runoff (Chiang and Chang, 2009) for finally predict the best choice of tomato cultivar, their type of production and their harvest date (Suarez et al., 2015), among others. Today, there are common ANN models, usable in artificial intelligence using programming libraries, recognized as efficient in many cases, and which do not do not require the in-depth development of a new optimal structure of the neural network. These methodologies therefore make it possible to focus on data analysis for classification (discrimination) and prediction (regression). [00731 Support Vector Machines - Support vector machines (SVM)
[0074] The SVM model uses the input data to construct a hyperplane (or hyperplanes) in high-dimensional space, to perform classification, regression or other tasks (RapidMiner, 2020a). In a classification problem, it is a question of determining the optimal separating hyperplane which maximizes the margin (distance between the hyperplane and the restricted subset of the samples closest to the hyperplane, samples also called "support vectors") . SVM models can be applied in many applications such as gear fault diagnosis (Xing et al., 2017) or to assess pavement condition (Hadjidemetriou et al., 2018). The LibSVM library of Chang and Lin (Chang and Lin, 2011; RapidMiner, 2020a) was used to develop the SVM models for the study of wines (Hsu et al, 2016)
[0075] Step (h): Preferably, the data processed in step (h) comprises: → at least one mineral analytical profile, advantageously metallic, of the alcoholic beverage, measured in step (d) from the following elements: * at least 5 mineral elements (called main) chosen from B, Na, Mg, P, S, Cl, K, Ca, * at least the following metallic elements (called trace metals): Fe; Cu; Zn; Min; * optionally at least one of the following metallic elements: Pb and Cd; * at least 10, preferably at least 30 and, even more preferably, at least 40 elements chosen from the following trace mineral elements: Rb, Cs, Sr, Ba, Ce, Ti, V, Cr, Co, Ni, Zr, Mo, Ag, Al, Ga, Sn, As, Br, I, Se; and / or from the following ultra-trace mineral elements: La, Pr, Nd, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb, Lu, Th, U, Sc, Y, Nb, Ru, Rh, Pd, Hf, Ta, W, Re, Os, Ir, Pt, Au, Kg, Tl, Bi, Sb; * possibly of at least part of the isotopes of these elements * measurements of the concentrations of these elements; * and / or the concentration ratios of these elements, and possibly of all or part of their isotopes; and at least 1, preferably at least 5 physicochemical parameters of the alcoholic beverage chosen from the group of parameters comprising - advantageously consisting of - TAV (Alcoholic Strength by Volume), Glucose + Fructose, AT (Total Acidity), Acetic Acid, SO 2 free, SO 2 total, pH, SO 2 active, Ethanal, Malic Acid, Lactic Acid, CO 2 , Tartaric Acid, Gluconic Acid, Glycenol, Optical Density, and all combinations of these parameters (preferably measured less than one week before or after bottling, or at a specified time relative to bottling).
[0076] Step (i): According to a first possibility, the use of the data processed in step (h) for the management and / or monitoring of agricultural production (factor *f 1 *) of the raw material for the manufacture of the alcoholic beverage, consists essentially of: * identify at least one parameter , ou une linear combination of these parameters, making it possible to improve the quality of the alcoholic beverage; * identify in the cultivation and / or harvesting processes one or more methods allowing to influence at least one parameter * modify the modality(ies) identified to change the parameter in the direction of improving the quality of the alcoholic beverage.
[0077] Step (I): According to a second possibility, the use of the data processed in step (h) for the management and / or monitoring of the manufacture of the alcoholic beverage essentially consists of: * identify at least one parameter - or a linear combination of these parameters, making it possible to improve the quality of the alcoholic beverage; * identify in the processes of vinification and / or assembly and / or aging, one or more methods allowing to influence at least one parameter * modify the modality(ies) identified to change the parameter in the direction of improving the quality of the alcoholic beverage.
[0078] Step (I): According to a third possibility, the use of the data processed in step (h) for the management and / or the monitoring of the conservation and / or maturation ( and / or consumption of this alcoholic beverage, essentially consists of: * identify at least one parameter - , or a linear combination of these parameters, making it possible to improve the quality of the alcoholic beverage; * identify in the processes of conservation and / or maturation and / or consumption one or more modalities allowing to influence at least one parameter * modify the modality(ies) identified to change the parameter in the direction of improving the quality of the alcoholic beverage.
[0079] Advantageously, the parameter is chosen from the group comprising - ideally constituted by -: the tasting parameters, preferably, balance, length, intensity, complexity, concentration and / or typicity; overall composition parameters, preference, alcoholic strength by volume, glucose & fructose, total acidity, acetic acid, SO 2 free, SO 2 total, pH, SO 2 active, ethanal, malic acid, lactic acid, CO 2 , tartaric acid, gluconic acid and / or glycerol; the parameters for the content of specific aromatic and / or coloring molecules.
[0080] Step (i): According to a fourth possibility, the use of the data processed in step (h) for the management and / or monitoring of the quality of the alcoholic beverage essentially consists in implement the method according to the invention in accordance with at least one of the 3 possibilities described in the preceding paragraphs.
[0081] Step (i): Within the framework of this fourth possibility, it can be envisaged in accordance with a remarkable variant of the invention, a use of the data processed in step (h) for the management and / or the monitoring of the quality alcoholic beverage, consisting essentially of: * identify one or more mineral profiles, preferably metallic, each constituting a specific target signature of a certain level of quality for an alcoholic beverage or of an alcoholic beverage promised at a certain level of quality; * search and select from a group of alcoholic beverages, those whose mineral profile, preferably metallic, corresponding to a target signature; * mark this or these selected beverages with an acquired or future quality guarantee label; ‘possibly use the mineral profiles, preferably metallic, of non-selected alcoholic beverages, to anticipate negative changes in these beverages and provide the necessary corrective solutions. The method according to the invention thus makes it possible to detect reliably, and possibly in advance, among alcoholic beverages, for example wines, during aging (barrel tank) and / or during maturation / conservation ( vat, barrel or bottle), real nectars or future nectars. The invention thus offers a screening means making it possible to select alcoholic beverages, for example wines of high quality, present or future. It is thus possible to grant selected alcoholic beverages, for example wines, labels guaranteeing present or future quality, which will have the consequence of mechanically increasing their commercial value.
[0082] Step (i): According to a fifth possibility, the use of the data processed in step (h) for the management and / or monitoring of the authenticity of the alcoholic beverage essentially consists To : ‘identify one or more metallic profiles constituting specific signatures of the origin of the alcoholic beverage; ‘ use this or these signatures as markers guaranteeing the authenticity of the alcoholic beverage; ‘ detect counterfeits using these markers.
[0083] Step (I): According to a sixth possibility, the use of the data processed in step (h) for the management and / or monitoring of traceability of the alcoholic beverage, essentially consists of: ‘identifying one or more metal profiles constituting specific signatures of the journey of the alcoholic beverage, from its origin to its consumption, including its manufacture, packaging and storage; * use this or these signatures as tracers of the alcoholic beverage, in particular in the context of investigations relating to food or health quality incidents. The method according to the invention thus gives access to traceability of alcoholic beverages, in particular wines, for the greatest benefit of the health and safety of consumers, from purchase to the end of the beverage's life. Thanks to this traceability, the consumer and the producer have a lot of information on the substances present in the drink as well as on the origin and on the conditions of production and delivery of this drink.
[0084] Step (ii): According to a seventh possibility, the use of the data processed in step (h) for the management and / or monitoring of the sale price of the alcoholic beverage, essentially consists of: ‘identify one or more metal profiles constituting (one) specific signatures of an average selling price of the alcoholic beverage in its category; * use this or these signatures as indicators of accessible selling price ranges and anticipate the future selling prices of the alcoholic beverage.
[0085] Step (il: Parameter ) gj. un p aram be linked to the methods of execution of the alcoholic beverage sector (e.g. wine), thanks to which the data processed in step (h), make it possible to implement corrective actions for these methods, to improve the quality of alcoholic beverage (e.g. wine).
[0086] In the case of wine and according to a preferred embodiment of the invention, this parameter | a Pa tasting 1 , the overall composition Pa 3 and the contents of aromatic molecules Pa 3 and / or specific dyes Pa 4 .
[0087] Pa 1 : Tasting (balance, length, intensity, complexity and typicality) ELlC(T) - BLIC(T) method in English - meaning: Balance - Length - Intensity - Complexity (and / or Concentration) - Typicality. Each parameter is added to the others in order to estimate the general quality of the wine.
[0088] Balance is fundamental. It is the main component of quality. We generally understand by balance of a wine a harmony between the different components of the texture in the mouth: acidity, alcohol and smoothness, tannins and sugars... It is often easier to define an unbalanced wine. It is a wine of which one of the components is in excess or insufficient in the wine. Thus, a wine that is too green with biting and acerbic acidity, or a wine that is too extracted with astringent and bitter tannins will be unbalanced wines. An unbalanced wine is generally judged as unsatisfactory (or mediocre) by tasters.
[0089] The Length of the wine is understood from the aromatic point of view, and it is preferable that the aromas of the wine remain in the mouth for a long time once it has been swallowed (or spat out in professional tasting). We often bet on a long wine in the mouth when we can feel for many seconds, or minutes, the aromas of a wine.
[0090] Intensity also has a link with the aromas of wine. The more fragrant the wine, the more intense it is said to be. Great wines are thus often very aromatic, although this is not always enough to make them wines of great quality (for example a cheap wine of gewurztraminer from Alsace, can be intense aromatically, but of a relatively short length by example).
[0091] Complexity is also a quality parameter. It expresses the number of aromas that we believe we detect when smelling the wine. The more the wine is rich in different aromas, the more it is considered to be of quality. A simple entry-level Muscadet from the Loire is always less complex than a Meursault from Burgundy. For some wines, perhaps still young, which have not yet revealed their full aromatic potential, we can then focus on the concentration of flavors. The richer a wine is, the more it is judged to be of high quality.
[0092] Finally, the Typicality (or Terroir) is the most complex notion to judge. The typicity of a wine can be understood as a unique and recognizable character of its place of production. Obviously, it is by dint of tasting wines from all regions of the world that one gradually builds up one's memory of the terroirs. A typical wine is a wine that inevitably has a “taste of the place”. In evaluating the quality of a wine, typicity often comes to the aid of fine and subtle wines, perhaps not as demonstrative and aromatic as certain "bombs" of the new world, made from from aromatic grape varieties... A Muscadet on lees, because of its terroir, is, for example, never very concentrated. On the other hand, it has this crystalline freshness, this saline minerality contributing to its typical and recognizable taste, in short, a great typicality.
[0093] Once we have evaluated the richness of each of the ELIC(T) parameters, we are able to better understand the quality of the wine. An unbalanced wine is a mediocre, even defective wine (if the smells are not clear or not pleasant). A simply balanced wine is considered acceptable. With an extra character (for example a lot of concentration), it is considered good; with two more characters, it is considered very good. And if all the parameters are present, then tasters can rate its quality as excellent. An excellent wine is therefore at the same time balanced, intense aromatically, with complex aromas, with a good length or even a certain typicality. [00941Pa 2 : Overall composition These are analytical data allowing the overall composition of a wine to be determined and to remain in the appropriate domain for good quality. These analytical data are chosen from the group comprising -advantageously consisting of-: TAV (Alcoholic strength by volume), glucose + fructose, AT (Total Acidity), acetic acid, SO 2 free, SO 2 total, pH, SO 2 active, ethanal, malic acid, lactic acid, CO 2 , tartaric acid, gluconic acid, glycerol and all combinations of these data.
[0095] The acquisition of these analytical data is carried out by the chemical analysis methods given in particular by the standards in force in the wine-growing field, such as for example: IRTF (Infrared spectroscopy with Fourier transform), automated visible spectroscopy, colorimetry , capillary electrophoresis, titrimetry, potentiometry...
[0096] Pa 3 : Aromatic molecules Some of these aromatic molecules have a positive effect on taste, others have a negative effect or are considered faults and / or contaminants, and others contribute greatly to the nose of wines. It is therefore possible, in accordance with the invention, to implement, with regard to the data processed in step (h), by statistical analysis and, possibly with the assistance of artificial intelligence, corrective actions aimed at: (i) to increase, and / or adjust and / or stabilize over time, the quantity of aromatic molecules with a positive taste effect; (ii) to reduce the quantity and break down and / or neutralize over time, the aromatic molecules with a negative taste effect or considered as defects and / or as contaminants; (iii) to adjust the quantity of aromatic molecules involved in the nose of wines.
[0097] Mention may be made, as examples, of aromatic molecules of type (i): Esters (acetaldehyde (fresh apple), isoamyl acetate (banana), ethyl acetate (acesant character), ethyl butyrate, ethyl isobutyrate , Ethyl 2-Methylbutyrate, Ethyl 2-Hydroxy-4-Mepentanoate, Ethyl Octanoate, Ethyl Decanoate, Ethyl Hexanoate, Ethyl Isovalerate, Isoamyl Acetate, 2-Phenylethanol, 3-Isobutyl -2-methoxypyrazine (IBMP) and isopropyl-methoxypyrazine (IPMP), sec-butyl-methoxypyrazine (SBMP)-green pepper and vegetable aromas-, Terpenes and Norisoprenoids (Terpenols: Geraniol, linalool, a-terpineol, nerol, citronellol, beta-damascenone, alpha-ionone, betaionone), Wood aromas (Trans- WhiskeyLactone, Cis-WhiskeyLactone (coconut, wood fresh), Eugenol, Isoeugenol (clove), Aceto-Vanillone, Vanillin, Ethyl-Vanillin, Ethyl-Vallinate (vanilla), Furfural, Methyl Furfural (toast, toasted almond), Methyl-guaiacol, Gaiacol (toast, smoky), Syringol, 4-Methyl-Syringol, 4-Allyl-Syringol, Acetosyringone (smoky), Syringaldehyde (smoky, vanilla undertones), O-Cresol (smoky, burnt), Trans-nonenal (green wood), Maltol ( caramel, cotton candy); → aromatic molecules of type (ii): • Molecules with strong odor power, mainly responsible for the corky or musty tastes present in wines: Volatile phenols (Ethyl-4-Phenol (Stable, leather), Ethyl-4-Gaïacol (spicy), Vinyl-4-phenol ( gouache, burnt rubber), Vinyl-4-guaiacol (clove sweet), haloanisoles [trichloroanisole (TCA), tetrachloroanisole (TeCA) and pentachloroanisole (PCA)], halophenols [trichlorophenol (TCP), tetrachlorophenol (TeCP) , pentachlorophenol (PCP) and tribromophenol (TBP)];
[0098] Geosmin (compound with strong odor power, which has an earthy smell - very marked moldy) • 3 molecules responsible for mouse taste: (popcorn, cooked rice, mouse urine, pet stores, etc.), 2-Acetyl-tetrahydropyridine (ATHP), 2-Acetyl-1-pyrroline (APY), 2- ethyl-tetrahydropyridine (ETHP); • mycotoxins [ochratoxin A (OTA)] and biogenic amines [Histamine, Methylamine, Ethylamine, Tyramine, Phenylethylamine, Putrescine, Isoamylamine, Cadaverine]; • benzaldehyde (benzoic aldehyde) with a bitter almond smell, and benzyl alcohol (benzyl alcohol from the plasticizer present in the epoxy resin coatings used in certain packaging for bottling), passes into the wine, where it is oxidized to benzaldehyde .) • 2-bromo-4-methylphenol (iodized tastes (of oysters sometimes)), ethyl carbamate, diethylene, monopropylene and monoethylene glycol; • molecules responsible for the smoke taste (aromas of cold ashes and an acrid character of the tannins), free forms: o-cresol, gaiacol, 4 methyl-guaiacol, Syringol, 4 methyl-syringol, 4 allyl-syringol or glycosylated precursors: glycosylated o-cresol, glycosylated gaiacol, 4-methyl-glycated gaiacol, gaiacol-glucopyranoside, gaiacol-gentiobioside, gaiacol-rutinoside, 4-methyl-gaiacol-rutinoside; → aromatic molecules of type (iii): Propan-1-ol, 2-methylpropan-1-ol, Isopentanols, 2-Methyl-butanol, 3-Methyl-butanol, Butan-1-ol, Butan-2-ol, But -2-ene-1-ol.
[0099] Pa 3 : Coloring molecules
[0100] It is possible, in accordance with the invention, to set up, with regard to the data processed in step (h), by statistical analysis and, possibly, with the assistance of artificial intelligence, corrective actions aimed at adjust the quantity and / or the evolution of the quantity over time, of coloring molecules acting on the appearance of the color.
[0101] Examples include: tannins and anthocyanins, resveratrol. The latter is a polyphenol present mainly in the skin of grapes. Resveratrol content depends on the grape variety (Pinot Noir, Grenache, Mourvèdre and Merlot contain more), vinification, geographical origin and exposure to cryptogamic diseases. This powerful antioxidant is said to have beneficial effects on human health.
[0102] Step (j):
[0103] Modalities of the cultivation and / or harvesting process making it possible to influence at least one preference parameter
[0104] In particular, this may consist of choosing the location of the vines, the type of vines, the type of crop, the type of harvest and finally the date of harvest and the state of maturity of the grapes at the time of the harvest.
[0105] The localization, for example with the GPS coordinates, of the plots of vines can be extended to the domains, appellation zones, town, regions. The location chosen also determines the relief [Plain - Slope - Steep slope (>20°) - Extreme slope (>30°)], the exposure / sunshine and the type of soil: Acid, basic, Clay - Limestone - Gravel - Pond - Chalk - Granite - Rolled pebbles - Schist - Silica - Silt - Sand - Gneiss - Sandstone - Others.
[0106] The type of vines is defined by the grape varieties (type and estimated distribution %), the age of the plans, the planting density.
[0107] The culture may be reasoned, conventional, organic, or biodynamic. It can also be defined in particular by the following elements: - Planting density (in feet / ha), - Fertilization of the vine (No - KMg binary fertilizers - Organic fertilizer - Mineral fertilizer): - Grassing (Not grassed - Grassed between rows on two - Grassed on all inter rows): - Other (spontaneous winter cover - sown winter cover - spontaneous permanent cover - sown permanent cover): - Phytosanitary control (No - Half dose - Diseases - Pests - Weedkillers) in particular quantity of copper spread / Ha: - Type of vine pruning [Goblet - Cordon de Royat - Simple Guyot - Double Guyot - Lyre - Mechanical precision pruning (TMP)]: - Irrigation of the vine (Across the furrow - Spraying on the foliage - Spraying close to the ground - Localized (drip)): - Origin of the water (if irrigation): Rain - Reserves - Groundwater - Rivers - Average age of the vine: number of years - Old vines - Young vines - Vintage: (and associated climate in the production area)
[0108] Harvesting or harvesting offers many adjustment variables, among others: - Harvest date: associated climate 1 week before and during; - State of maturity of the grapes at the time of harvest and in particular the often analyzed maturity control parameters: Density, Glucose + Fructose, Probable Degree, Acquired TAV, AT, Volatile Acidity, pH, Malic, Tartaric, Gluconic Acids, Citric, Glycerol: Indicators of health status, Potassium, Ammoniacal nitrogen, a-amino nitrogen, Total assimilable nitrogen, Delta C 13 index: Indicator of water stress. - Type of harvest: Manual - mechanical.
[0109] Step (i):
[0110] Modalities of the vinification and / or blending and / or aging process making it possible to influence at least one preference parameter
[0111] These methods relate to several stages in the manufacture of the alcoholic beverage that is wine, corresponding to a particular mode of implementation of the process according to the invention, namely: reception of the harvest and prefermentation operations, alcoholic fermentation and fermentation operation, vatting and alcoholic fermentation, aging and operation following vinification, decision to end vinification and follow-up control of alcoholic and malolactic fermentation.
[0112] Reception of harvest and preferential operation - Sorting upon receipt of the harvest: manual - mechanical; - Destemming: type of scraper; - Crusher: pre-fermentation crusher made (to release the juice more easily); - Press (pressing berries for vinification into white wine) types of presses (mechanical - hydraulic - pneumatic - horizontal - vertical); - Marc pump (large auger) type and materials of the auger; - Wine tank material used (stainless steel - concrete - wood - ceramic); - Tank volume and filling rate (important for oxygenation); - Maceration time of the skins and pips in the must: for example in number of days; - Addition after tank filling - oenological corrections: sulphiting, chaptalization, concentration by reverse osmosis, acidification, deacidification, dealcoholization, enzymation, wood chips; -Extraction and / or addition of juice (juice can be removed in order to increase the grape skin / juice ratio, in order to obtain more concentrated wines, bleeding is then used for rosé or a less qualitative batch); - Management of clarification and aromatic properties: possible stabling (for white and rosé vinification) increases the expression of aromas; - Settling (clarifying the must by removing suspended particles and various impurities, especially for whites and rosés); - possible blending of different musts (in the case of vinification of different grape varieties, plots, etc.); - pellicular maceration (for direct pressing white and rosé vinifications; - contact of the skins and the must for a few hours to diffuse the primary aromas and anthocyanins (for the rosés).
[0113] Alcoholic fermentation - fermentation operations - Operations prior to vatting: carbonic maceration (essentially for primeur wine) from whole grapes (not destemmed and not crushed) allowing alcoholic fermentation to begin in the grape berry; - Cold pre-fermentation maceration (5-15°C) with whole grapes allows alcoholic fermentation to begin in the grape berries.); - Crushing (Crushing can take place before actual vinification or after); - Possibly, pellicular maceration [this operation also allows a natural yeasting from the indigenous yeasts present on the skins of the berries of the grape (which reinforces the typicality of the wine)]; - Carbonic maceration or cold prefermentation maceration.
[0114] Vatting - alcoholic fermentation - Pure indigenous yeasts or added yeast (selected by laboratories with high fermentation potential); - Temperature and thermoregulation time of the tanks; - Daily reading of must density (sugar denser than alcohol) and possible additions; - Possible techniques for improving & extracting aromas, color and tannins: pigeage (between 8 and 20 days depending on the wine) consisting of pushing the cap of marc into the liquid part of the must in fermentation while crumbling to promote the diffusion of phenolic compounds and aromas.); - Pumping over (recover the must during fermentation accumulated in the bottom of the tank to pour it over the cap of marc which floats on the surface of the tank); - Load shedding consisting of recovering all the must during fermentation accumulated in the bottom of the tank and transferring it to a second tank. It is then returned to the cap of marc which has compacted and is drained to the bottom of the vat to improve maceration), possible hot maceration of the marc of the vat by dissociating it temporarily (1 to 2 days) from the rest of the vintage; - Possibly hot maceration, mutage (for specific wines stops fermentation to obtain a natural sweet wine, possible clicking (or macro-oxygenation) allowing a short and punctual amount to be added), aromatization... - At the end of fermentation the juice is composed of a large quantity of alcohol, which accentuates the extraction and in particular that of undesirable compounds, the extraction will then be limited to the minimum, we simmer. The cap of marc is now only lightly watered (very little pumping over) daily, or even every two days, in order to renew the juice present in the marc and to prevent it from getting pricked. At this time, the tasting is essential, when the cellar master and the oenologist consider that the material and the fat have been extracted enough, the tank is drained to draw wine with bitterness, greenness and / or of drought. The duration of vatting varies according to the quality of the grapes and the desired wine, it generally varies from 10 to 30 days.
[0115] - The malolactic fermentation generally takes place after the alcoholic fermentation. It reduces the acidity of certain wines. - Sulfiting the wine protects it from oxidation and subsequent potential microbiological deviations, it occurs after malolactic fermentation so as not to hinder it. - The aging of the wine follows with possible micro-oxygenation and / or macro-oxygenation. Aging can be done in different ways: aging in vats - aging the wines in wood - type of vat - time and conditions; wine aging with micro-oxygenation and / or macro-oxygenation: oxygenation conditions - Eventually, are finally implemented, before bottling, operations: • clarification and stabilization. • Tartriue stabilization. • filtration [Classic cellulose plate or soil filter - Tangential micro-filtration system (MFT)], • and / or fining [Bentonites - Fish glue - Casein - Egg albumin - Gelatin - PVPP - Soils and silica gels, ...]. Some of the correction operations of the method according to the invention in particular consist in acting on: • the fining of white and rosé wines (protein stabilization fining, stabilization fining and clarification, ferrocyanide treatments), * the stability of white and rosé wines (Pinkink of white wines, protein stability, tartaric stabilization, treatments for tartaric stability), * and / or red wine fining (stabilization and clarification fining, tartaric stability, clogging index and filtration behavior). At the end of the vatting, the flow by gravity of the free run wine, the devatting and then the pressing of the marc, give the press wine. A blend can be made at any time, between free run and press wine, between two cuvées, between varietal wines to make a blended wine.
[0116] End of vinification decisions (analysis)
[0117] Control and monitoring of alcoholic and malolactic fermentation is based on the following indicators: Glucose + Fructose, Acquired TAV, Probable Degree, AT, Acetic Acid, SO 2 free, SO 2 total, pH, Malic Acid, Lactic Acid,
[0118] An assessment is made after racking, on the basis of the following indicators: SO 2 total, pH, SO 2 active, Ethanal, Malic Acid, Lactic Acid, CO 2 , Tartaric acid, Gluconic acid, Glycerol, Iron, Copper (whites and rosés) - Tasting - OTA index, Color intensity on request),
[0119] When bottling or in BIB, we dose: TAV, Glucose + Fructose, AT, Acetic acid, SO 2 free, SO 2 total, pH, SO 2 active, Ethanal, Malic Acid, Lactic Acid, CO 2 , Tartaric acid, Iron, Copper, 2 protein stability tests (whites - rosés), we taste, we carry out a microbiological analysis by cytometry, and, possibly a cold test and / or a clogging index on request.
[0120] Step (i): According to a remarkable embodiment of the invention, step (I) of the method can comprise a sub-step (i c ) consisting of an action to correct the mineral profile which may consist of: - extract metals before bottling using filtration, preferably cross-flow filtration associated with metal-complexing polymers (put on the other side of the membrane with respect to the wine); - and / or extracting metals at a stage before bottling or after opening the bottle, by trapping, preferably using a biopolymer gel functionalized by ultra-chelating agents directly placed in contact with the wine.
[0121] The method according to the invention takes the advantage of focusing on the mineral / metallic analysis of alcoholic beverages, in particular wines, by using suitable containers for the collection of samples, by processing the data statistically, preferably by aiming learning processes based on artificial intelligence and by collecting the data collected in a database provided for this purpose. This makes it possible to know in particular the impact of metals on the tastes of alcoholic beverages (e.g. wines), and in doing so to introduce analytical objectivity into all ratings, classifications and oenological competitions or other professional alcoholic beverage competitions. Alcoholic beverages, in particular wines, can thus be distinguished from each other, according to their quality at a time t, but also by taking into account the prospects for the evolution of this quality, in view of the mineral profile (e.g. metallic) of the alcoholic drink. The fact of having access, thanks to the process according to the invention, to the evolutionary potential of the alcoholic drink, is a very interesting new criterion, which facilitates its evaluation and makes it possible to assign scores with greater fairness and impartiality. The rapid analysis of the data in the database using the methods and tools according to the invention constitutes an effective aid for informed, and above all objective, decision-making for the classification or evaluation of alcoholic beverages, in particular wines. , but also upstream to provide all desirable preventive and / or corrective actions at all stages of the production of alcoholic beverages.
[0122] DEVICE
[0123] According to a second of its aspects, the invention relates to a device for implementing the method according to the invention, characterized in that it comprises a sample library comprising at least one enclosure in which the samples collected at the step (a) in inert containers each closed by a stopper and in that this chamber is suitable for placing these samples under given temperature, pressure, humidity and atmosphere conditions.
[0124] The invention also relates to a computerized management system for the sample library and the database. Description of figures
[0125] The appended figures illustrating non-limiting examples of embodiment, and in which:
[0126] [Fig. 1] Figure 1 is a distribution of the wines analyzed in the examples, according to their grape varieties.
[0127] [Fig. 2] Figure 2 is a distribution of the wines analyzed in the examples, according to their origins - their regions.
[0128] [Fig. 3] Figure 3 is a distribution of the wines analyzed in the examples, according to their cultivation methods.
[0129] [Fig. 4] Figure 4 is a graph in which: - the ordinate corresponds to the number of wines analyzed in the examples which fall into a price category and which are counted, - and the abscissa corresponds to the Cl concentration expressed in μg / L.
[0130] [Fig. 5] Figure 5 is a graph in which: - the ordinate corresponds to the number of wines analyzed in the examples which fall into a category (red wine or white wine) and which are counted, - and the abscissa corresponds to the concentration in K expressed in μg / L.
[0131] [Fig. 6] Figure 6 is a graph in which: - the ordinate corresponds to the number of wines analyzed in the examples which fall into a price category and which are counted, - and the abscissa corresponds to the concentration in K expressed in μg / L.
[0132] [Fig. 7] FIG. 7 is a graph representing the distribution of the Mg concentrations of wines analyzed in the examples, as a function of the quality of the wine (here translated by a selection of the scores obtained).
[0133] [Fig. 8] FIG. 8 is a graph representing the distribution of the K concentrations of wines analyzed in the examples, as a function of the quality of the wine (here translated by a selection of the scores obtained).
[0134] [Fig. 9] FIG. 9 is a graph representing the distribution of the Ca concentrations of wines analyzed in the examples, as a function of the quality of the wine (here translated by a selection of the scores obtained).
[0135] [Fig. 10] FIG. 10 is a graph representing the distribution of the Na concentrations of wines analyzed in the examples, as a function of the quality of the wine (here translated by a selection of the scores obtained).
[0136] [Fig. 11] Figure 11 shows a statistical correlation matrix between the mineral elements analyzed (lanthanides and W, S, Nb).
[0137] [Fig. 12] Figure 12 presents the mineral profile of a wine from the 2 ème series of examples, according to a star graphic representation (Kiviat diagram).
[0138] [Fig. 13] Figure 13 is a graph giving the distribution of metallic and mineral elements on the different concentration classes of wines from the 3 ème series of examples
[0139] [Fig. 14] to [Fig 17] Figures 14 to 17 show the mineral profiles represented according to Kiviat diagrams of red, white, rosé and sparkling wines from the 3 ème series of examples. EXAMPLES 1 ère series of examples: §
[0140] to
[0220]
[0140] Collection [step (a)] and storage of samples [step (c)J.
[0141] The samples are taken directly from 24 bottles of commercial wine: -1- Saint Joseph - Saint Ronain Cellars - Jacques Delorme -2- Macon Villages - Graubünden area -3- Hermitage - E. Guigal -4- Pays d'Oc - Naturae Chardonnay - Gerard Bertrand -5- Veneto - Canteen Maschio - Verduzzo -6- Saint Emilion Mountain - Chateau Petit Clos du Roy - François Janoueix -7- Saint Emilion Mountain - Chateau Petit Clos du Roy (Premium) - François Janoueix -8- Pomerol - Château l'Évêché - François Janoueix -9- Saint Emilion grand cru - Château Condat - François Janoueix -10- Meursault - Domaine Saint Marc - Under the Velle -11- Burgundy Aligoté - Nuiton Beaunoy -12- Burgundy - Nuiton Beaunoy - Pinot Noir -13- Mâcon - The Grison area - Mâcon Péronne -14- Côtes du Rhône - E. Guigal -15- Chassagne Montrachet - Domaine Paul Pillot - 1er Cru Les Champs Gains -16- Côte du Rhône - Cellar of Dolphins - Old Vines -17- Montagny - Domaine de la Guiche - Montagny 1er Cru -18- Chassagne Montrachet - Jean Bouchard -19- Côte-Rôtie - Domaine Vemay - Gisele -20- Bourgogne Passe tout grains - Bernard Millot - 100 Names -21- Nights Saint Georges - Domaine Henri GOUGES - Red Villages -22- Maranges - Michel Sarrazin and Son - Côte de Beaune -23- Givry (red Burgundy) - Michel Sarrazin et Fils - Givry 1er cru - The old vines -24- Gevrey Chambertin (Red Burgundy) - Domaine Denis MORTET - My Five Terroirs
[0142] Once the bottle has been uncorked, part of the wine is emptied (at least 100 ml and less than half of the contents of the bottle), then two "metal free" storage bottles (ultra-low tube MetalFree™ Centrifuge Tubes 15 ml ref 3134-345 - labcon®) are filled with wine (at least 10 ml per bottle). This step is done less than an hour after opening the bottle. The felt of pre-emptying part of the bottle avoids surface deposits and washes the flow area of dust that could be in contact. The felt of leaving at least half of the bottle before sampling minimizes the sampling of deposits at the bottom of the bottle.
[0143] The sample tubes are then placed in a storage area at room temperature.
[0144] A part of the content of one of these sample tubes is taken, to carry out analyzes of metal content by ICP-MS (see below step (d)_§
[0143] -
[0149] ).
[0145] [step (b)] Each tube is referenced and the available information about the wine is entered into a data table. This information is listed below in the passage relating to step (e) creation of the database.
[0146] [step (d)] Analysis of the content of mineral elements and trace metals
[0147] The reagents used are of ultra-pure quality: HNO 3 35% (Suprapur, Merck), 30% HCl (Suprapur, Merck) and doubly deionized water (resistivity 18.2 MQ.cm) dispensed via a purification system (Pure La b, Elga)
[0148] For each wine, a 400 pL aliquot is taken and diluted to 1 / 10 with 1% nitric acid in a Wheaton polypropylene vial 15 x 45 mm (diameter x height) then placed in an Agilent integrated automatic sample changer ( the ace).
[0149] The analysis is carried out in semi-quantitative mode using an ICP-MS Agilent® 7700. This is equipped with an octopole which will be used in helium collision mode (He flow rate = 4.3 mL / min) before the entry into the quadrupole type mass analyzer. Each sample is then nebulized using a Micromist® nebulizer then introduced into the ICP-MS after passing through a Scott chamber, cooled to 2°C. An equilibration and signal stabilization time of 60 s is programmed before proceeding. to the actual measurement. The analysis is then carried out over the entire range of masses (7 to 238) at the rate of 6 acquisition points per mass and 0.1 s per point.
[0150] Obtaining semi-quantitative data is ensured by the use of one-point calibration using a multi-element standard (5 ppb / element). This was prepared from 1000 ppm monoelemental solutions of Li, Mg, Co, Y, Ce, Tl (Plasma CAL ICP-MS) diluted in 1% nitric acid.
[0151] After each analysis, the needle of the passer is rinsed 30 s with water then 30 s with 2% HCl and finally, 30 s with HNO 3 1%.
[0152] Between each sample, a blank sample (HNO 3 1%) is injected to ensure the absence of cross-contamination.
[0153] The results for the 24 wines measured are presented in tables 1 to 40 (see below). Between different analyses, reference solvent measurements are regularly carried out to look at the relevance of the signals obtained. On the different elements studied, several mineral elements stand out and give usable measurements for certain samples, representative for integrating into a sheet of metal profile.
[0154] There are thus 60 elements that can be directly integrated into a wine metal profile sheet: Li, B, Na, Mg, Al, Si, P, S, K, Ca, Tl, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, Ga, Ge, As, Br, Rb, Sr, Y, Zr, Nb, Mo, Rd, Cd, In, Sn, Sb, I, Cs, Ba, La, Ce, Pr, Nd, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb, Lu, Hf, W, Pt, Hg, Tl, Pb, Bi, Th and U.
[0155] [Table 1]
[0159] [Table 5]
[0176] To verify the relevance of the measurements, dilution tests were carried out on a wine sample (ref Wine N°20) at 3 levels: dilution by 20, by 10 and by 5 in nitric acid.
[0177] Semi-quantitative analysis was performed on the main elements. For each element, the measurements obtained were compared (dilution value 5 / dilution value 10) and (dilution value 10 / dilution value 20). When for these two values, the ratio is equal to 2+-0.5, it can be considered that the measurement of the element is reliable in the domain and that the element can be considered as reliable to use. We thus found 50 reliable elements to measure for this sample: Na, Mg, Al, S, K, Ca, Sc, V, Cr, Mn, Fe, Co, Ni, Cu, Ge, Rb, Sr, Y, Zr, Mo, Pd, I, Cs, Ba, La, Ce, Pr, Nd, Eu, Gd, Tb, Dy, Tm, Yb, W, Tl, Pb, Bi, Th and U.
[0178] When one value of the ratio is equal to 2 +-0.5 and the other value 2+-1 , the element can be measured more specifically by adjusting the dilution setting, and can be considered interesting to interpret. In this case, 11 additional elements were found for this sample: Si, Ti, Zn, Ga, Br, Nb, In, Sn, Sm, Ho, Er.
[0179] And, finally 2 elements whose two ratios are equal to 2+-1: Cd, Hf
[0180] It is noted that more than 60 mineral elements can thus be measured by semi-quantitative ICP-MS on this sample and entered in a sheet of metallic profile of the sample. See tables 22 to 40 below.
[0181] [Table 22] complementary physico-chemical analyses, and finally the qualitative data on the taste and quality of the wine, or the feeling via the notes and the selling prices.
[0203] The following parameters, if available, are collected and recorded in the table for each wine collected: Origin / AOC / AOP / PGI / Name of the wine estate - Other names (variety, cuvée, etc.) / Type / Vintage / Cultivation method / % alcohol / Selling price / Sulphites / Domain website and Domain contact / Time aging / Country / Region / Vi ti col e region / GPS coordinates / Appellations / Type of glass Capacity (in ml) / Cork / Storage / Opening date / Opening location / Sight / Nose / Taste / Climate / Exposure-precise orientation / Influence of wind / Relief / Altitude / Origin of watering / Irrigation of the vine / Type of soil / Types of cultivation / Area in Ha / Planting density / Fertilization of the vine / Grass cover / Phytosanitary control / Type of vine pruning / Average age of the vine / Type of formation / Picking / Sorting / Destemming / Types of presses / Maceration time of the skins and pips in the must) / Wine vat material / Wine vat volume (in hl) / Addition of yeast / Filling & Clarification / Filtration / Blending / Review notes: wine experts / public (Vivino, Robert Parker (Wine advocate), Wine Spectator, Decanter, Jancis Robinson, James Suckling, JEB DUNNUCK, Vinous, Hachette , Wine EnthusiastyLabel / Certifications / Medals:
[0204] For the wines analyzed, the appended figures 1 to 3 show, by way of examples, the distribution obtained according, respectively, to information on the grape varieties, the origins - regions and the cultivation methods.
[0205] [Step (h)] Preliminary statistical studies
[0206] To analyze the data, we use, in this example, the python language with the numpy, pandas, seabo and matplotlib libraries (non-exhaustive list). The working method begins with a first data processing by analyzing the basic properties of all the data. Then comes the phase of viewing certain characteristics according to classes defined beforehand (for example, grouping according to price ranges, rating ranges, etc.) - Loading data in Python The following libraries are imported under Python: pandas, seabo (for graphical representation), numpy, matplotlib and scipy.stats. The excel files are loaded in the form of Pandas DataFrame: the first contains the concentrations of metals, the second the information relating to the origins, descriptions and quality of the wine. These files are merged to have a single DataFrame, grouped using the common information: “Sample Name”. - Creation of DataFrame subset
[0207] Seaborn is used to create histograms of quantitative variables with the seabo function. histplot. We superimpose on the histogram the KDE option of seaborn. It is a representation of the data using the probability density curve.
[0208] Seabom.plot.pie is then implemented to represent qualitative variables such as origin, type of wine, method of cultivation etc.
[0209] In order to better represent the data graphically, subsets of interest are created. Examples of subsets studied are: metals by category, prices (by setting lower and upper bounds) and Vivinos ratings.
[0210] These subsets are contained in new Pandas DataFrames. - Representations of quantitative variables It is therefore now possible to draw histograms of metals for each subset, namely, the histograms of the concentrations of metals belonging to the subsets price, then type of wine and finally note vivino.
[0211] Once a relationship between two variables has been identified (for example the price and the Vivino rating), it is possible to use seaborn. Implot to visually confirm the relationship.
[0212] - Correlations To establish a quantitative correlation, we use the function , to be used in the form: DataFrame.corr(). We thus obtain a correlation between the price and the Vivino Rating of 0.75, which proves that the variables are correlated. This information can be found thanks to the “seaborn. clustermap” which groups the most correlated metals into dusters. This graph is symmetric, and the values of correlations on the diagonal are equal to 1, since it is the correlation between the metal and itself. By plotting this graph for metals of “ultra-trace” types, we notice strong correlations between the metals La, Ce, Pr, Nd, and Eu.
[0213] -Additional scans Finally, a graph used here but which is relatively overloaded is the seabom.pairplot. It is about the abscissa of the variables xi Xn as a function of the ordinates of the variables xi,...x n-i . Id, these are the main metals. On the diagonal, there is simply the histogram of the metal Xn. It is then possible to look visually for correlations. Figures 4, 5 and 6 are examples of graphs obtained, in which: - the ordinate corresponds to the number of wines falling into the category (price for figures 4&5 and red wine or white wine for figure 6) and counted, - and the abscissa corresponds to the concentration (in Cl for figure 4 and in K for figures 5&6) expressed in μg / L.
[0214] The graph in Figure 4 shows that the chlorine content increases with the price of wine.
[0215] The graph in figure 6 shows a difference between white and red wine on the potassium level.
[0216] The 4 graphs of figures 7 to 10 representing the distribution of the concentrations of Mg, K, Ca and Na, respectively, according to the quality of the wine (id translated by a selection of the scores obtained). We distinguish a profile that is emerging for quality wines, rated more than 4, which have preferential contents for the 4 elements Mg, K, Ca and Na centered, respectively, around 650,000, 200,000, 3,250 and 5,000 μg / L.
[0217] [step (i)] Use of statistically analyzed data
[0218] From the different measurements performed on the ultra-traces, we obtain a correlation matrix shown in Figure 11.
[0219] The scale on the right in Figure 11 corresponds to the Pearson coefficient (the value -1 means that the variables have inversely correlated, +1 linear correlation and zero no correlation). It is calculated according to the expression covariance of the matrix (X, Y) divided by std(X)*std(Y) (standard deviation of X times standard deviation of Y): cov(X,Y) / std(X) std(Y)Pearson's coefficient is an index reflecting a linear relationship between two continuous variables. The correlation coefficient varies between -1 and +1, 0 reflecting a zero relationship between the two variables, a negative value (negative correlation) meaning that when one of the variables increases, the other decreases; while a positive value (positive correlation) indicates that the two variables vary together in the same direction. We see a very clear correlation, for example, at the level of the lanthanides between them, and also associated quite strongly with W, As and Nb. 2 ème series of examples: §
[0221] to
[0226]
[0220] The samples are taken from a bottle of commercial wine AOC Saint Joseph Domaine The Pizon 2020 Fabrice and Loïc Peychon.
[0221] The protocol described for the first series of examples in paragraphs
[0140] to
[0150] is implemented, with the difference that, during sampling, the wine is stored after dilution to one third in nitric acid (HNO3) ultra-pure at 1% with addition of an internal indium standard at 10 μ / gL. The use of an acid solution such as HNO3 stabilizes the wine by limiting the problems of mineral precipitation and adsorption on the walls of the containers. On the other hand, the use of an internal standard makes it possible in particular to control any evaporation during storage. The samples are again diluted by 5 in 1% HNO3 just before the analyses. These latter are conducted semi-quantitatively and in collision mode with helium.
[0222] Table 41 below shows the results obtained:
[0223] [Table 41]
[0224] Figure 12 shows the mineral profile of the wine submitted for analysis, in the form of a Kiviat diagram which presents the results in deciles and with reference to all of the 590 red wines from France analyzed under the same conditions.
[0225] Commentary on figure 12: a specific profile of the wine appears, showing for example compared to all red wines, a very high relative concentration of Ni and Cs (last deciles) and a high content of magnesium and silicon (before last decile), the wine is very low in sodium and aluminium. 3*"* series of examples: §
[0227] to
[0230]
[0226] The samples of 1,200 commercial wines produced in France, of types, regions, estates, AOCs and various vintages, are taken and analyzed in the same way as for the second series of examples.
[0227] The analyzes conducted as part of this study on 1,200 wines produced in France perfectly illustrate the distribution of metallic and mineral elements over the different concentration classes (figure 13).
[0228] The results of analyzes carried out on these 1200 wines give mineral profiles represented according to Kiviat diagrams in figures 14 to 17, in which: * Figure 14: mineral profile (dark curve) associated with the average of n=587 red wines compared to a mineral profile (light curve) associated with the average of (1200 - 587) non-red wines. Figure 15 Mineral profile (dark curve) associated with the average of n=323 white wines, compared to a mineral profile (light curve) associated with the average of (1200 - 323) non-white wines. Figure 16: mineral profile (dark curve) associated with the average of n=189 rosé wines compared to a mineral profile (light curve) associated with the average of (1200 - 189) non-rosé wines. Figure 17: mineral profile (dark curve) associated with the average of n=110 sparkling wines compared to a mineral profile (light curve) associated with the average of (1200 - 110) non-sparkling wines.
[0229] These profiles make it possible to assign a real mineral signature to the wines. It clearly appears, on average representations of the categories of red, white, rosé and sparkling wines (Figures 14 to 17 respectively), a clear differentiation of their mineral compositions, compared to those of other wines.
Claims
Demands
1. A method, including computer-based methods, for managing and / or monitoring at least one factor *f x * chosen from a set of factors including: *f 1 * the agricultural production of a raw material for the manufacture of an alcoholic beverage, preferably wine, *f 2 * the manufacture of this alcoholic beverage; *f 3 * the preservation of this alcoholic beverage; 'f 4 * the maturation of this alcoholic beverage; *f 5 * the consumption of this alcoholic beverage *f 6 * the quality of this alcoholic beverage; *f 7* the authenticity of this alcoholic beverage in relation to a reference chosen from the group comprising -advantageously constituted by-: the names of the wine and the estates, the appellations of origin; the geographical indications; the traditional specialities guaranteed; the labels; the registered trademarks; and their combinations; *f8* the traceability of this alcoholic beverage; *f9* the selling price of this alcoholic beverage; this process essentially consists of: (a) collect at least one, preferably at least two samples of the alcoholic beverage, place each of them in a container inert to the alcoholic beverage and seal the latter tightly with an equally inert stopper; (b) assign to each sample data on the alcoholic beverage, chosen from the group comprising -advantageously constituted by- data on origin, data on terroir, data on manufacture, data on preservation and maturation, data on consumption, physico-chemical data, qualitative data, in particular organoleptic data, economic data, commercial data, and combinations of these data; (c) possibly retain at least part of the samples collected in step (a), under specified conditions; (d) analyze each sample to determine at least one mineral profile, preferably metallic; (e) establish a database relating to the samples and from step (b) and step (d); (f) possibly, complete and / or update the data assigned in step (b), at least once, on all or part of the samples; (g) possibly, complete and / or repeat the analyses carried out in step (d), at least once, on all or part of the samples from step (c); (h) process these data by statistical analysis, advantageously by automatic or semi-automatic methods based on statistical processing, and, even more advantageously by data mining, preferably using artificial intelligence tools and / or other data mining techniques; (I) and use the processed data for the management and / or monitoring of at least one of the factors *f x * mentioned above.
2. The method according to claim 1 characterized in that the number N of alcoholic beverages collected at step (a) is such that, in ascending order of preference: N ≥ 500; N ≥ 1,000; N ≥ 10,000.
3. A method according to at least one of the preceding claims, characterized in that: * Origin data includes the name of the alcoholic beverage, the name of the producer, the name of the estate, the year of production, the date of sampling, the name of the vintage, the batch number, and / or the type of alcoholic beverage; * terroir data includes the designation of origin, geographical indication, country, region, place, plot, grape varieties, distribution of grape varieties, exposure, sunshine, planting density (in vines / ha), type of vine pruning, cultivation method, vine fertilization, cover cropping, plant protection, watering, average age of the vine, relief, soil type, water origin, and / or vine irrigation; * Manufacturing data includes the type of harvest, the date of harvest, the type of sorting and destemming, the type of vinification, the type of press, the maceration time of the skins and seeds in the must, the material of the tanks, the addition of yeast or not, the type of fining and clarification, the filtration system, and / or blending; * Storage data includes storage time, successive container types, capacity, temperature, humidity, type of cap, and / or packaging date; * Consumption data includes the presence and level of sulfites, the presence and level of phenolic compounds, the percentage of alcohol, and / or the presence and level of aromatic compounds; * Qualitative data, particularly organoleptic data, include assessments of Balance, Length, Intensity, Complexity and Typicality [ELIC(T) method], color, aromas, tastes, and / or the duration of aroma expression in the mouth, preferably expressed in the tail * Economic data includes price, sales volume, and / or sales amount; * Commercial data includes labels, awards / medals in competitions, rankings, and / or scores received.
4. A method according to at least one of the preceding claims, characterized in that the mineral profile analyzed in step (d) for the samples comprises: * at least 5 mineral elements (called principal elements) chosen from B, Na, Mg, P, S, Cl, K, Ca, * at least the following metallic elements (known as trace metals): Fe; Cu; Zn; Mn; * possibly at least one of the following metallic elements: Pb and Cd; * at least 10, preferably at least 30 and, even more preferably, at least 40 elements chosen from among the following trace mineral elements: Rb, Cs, Sr, Ba, Ce, Ti, V, Cr, Co, Ni, Zr, Mo, Ag, Al, Ga, Sn, As, Br, I, Se; and / or from among the following ultra-trace mineral elements: La, Pr, Nd, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb, Lu, Th, U, Sc, Y, Nb, Ru, Rh, Pd, Hf, Ta, W, Re, Os, Ir, Pt, Au, Kg, Tl, Bi, Sb; * possibly of at least some of the isotopes of these elements * measurements of the concentrations of these elements: * and / or the concentration ratios of these elements, and, possibly, of all or part of their isotopes.
5. A method according to at least one of the preceding claims characterized in that step (d) comprises the analysis of chemical and physical parameters of the alcoholic beverage, these parameters being, preferably, chosen for an alcoholic beverage consisting of wine, in the group comprising -advantageously consisting of-: ABV (Alcoholic strength by volume), Glucose + Fructose, TA (Total Acidity), Acetic Acid, Free SO2, Total SO2, pH, Active SO2, Ethanal, Malic Acid, Lactic Acid, CO2, Tartaric Acid, Gluconic Acid, Glycerol, optical density (absorbance at one or more wavelengths 280, 420, 520 and 620 nm) and all combinations of these parameters.
6. A method according to at least one of the preceding claims, characterized in that the data processing according to step (h) essentially consists of → using at least one of the following means: * Exploratory analyses and logistic regressions to perform classifications based on Principal Component Analysis (PCA) * discriminant analyses (LDA or Linear Discriminant Analysis in English) * predictive model algorithm, preferably chosen from the group comprising - ideally consisting of -: Random Forest (RF) Decision Trees and / or Artificial Neural Networks (ANN) and / or Support Vector Machines (SVM); → to perform at least one of the following actions: * Identify and predict the origins of the different mineral elements within a wine; * Identify and predict the impacts on the quality of a wine and the evolution of the quality of a wine of the different mineral elements within a wine; * Adjust the origins to adapt the quality.
7. A method according to at least one of the preceding claims, characterized in that the data processed in step (h) includes: → at least one mineral analytical profile, advantageously metallic, of the alcoholic beverage, measured in step (d) from the following elements: * at least 5 mineral elements (called principal elements) chosen from B, Na, Mg, P, S, Cl, K, Ca, *at least the following metallic elements (called trace metals): Fe; Cu; Zn; Mn; * possibly at least one of the following metallic elements: Pb and Cd; * at least 10, preferably at least 30 and, even more preferably, at least 40 elements chosen from among the following trace mineral elements: Rb, Cs, Sr, Ba, Ce, Tl, V, Cr, Co, Ni, Zr, Mo, Ag, Al, Ga, Sn, As, Br, I, Se; and / or from among the following ultra-trace mineral elements: La, Pr, Nd, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb, Lu, Th, U, Sc, Y, Nb, Ru, Rh, Pd, Hf, Ta, W, Re, Os, Ir, Pt, Au, Kg, Tl, Bi, Sb; * possibly of at least some of the isotopes of these elements * measurements of the concentrations of these elements; * and / or the concentration ratios of these elements, and, possibly, of all or part of their isotopes; → and at least 1, preferably at least 5, physicochemical parameters of the alcoholic beverage chosen from the group of parameters comprising - advantageously consisting of - TAV (Alcoholic Strength by Volume), Glucose + Fructose, AT (Total Acidity), Acetic Acid, Free SO2, Total SO2, pH, Active SO2, Ethanal, Malic Acid, Lactic Acid, CO2, Tartaric Acid, Gluconic Acid, Glycerol, optical density, and all combinations of these parameters.
8. A method according to claim 6 characterized in that the data processed in step (h) is used for the management and / or monitoring of the quality of the alcoholic beverage, essentially consists of: * identify one or more mineral profiles, preferably metallic, each constituting a specific target signature of a certain level of quality for an alcoholic beverage or of an alcoholic beverage promised to a certain level of quality; * search and select from a group of alcoholic beverages, the one(s) whose mineral profile, preferably metallic, corresponds to a target signature; * mark this or these selected beverages with a quality guarantee label, whether acquired or to be acquired; 'possibly use the mineral profiles, preferably metallic, of unselected alcoholic beverages, to anticipate negative developments in these beverages and provide the necessary corrective solutions.
9. A method according to at least one of the preceding claims, characterized in that the data processed in step (h) is used for the management and / or monitoring of authenticity (factor *f 7 *) of the alcoholic beverage, essentially consists of: 'identify one or more metallic profiles constituting specific signatures of the origin of the alcoholic beverage; * use this signature or these signatures as markers guaranteeing the authenticity of the alcoholic beverage; * detect counterfeits using these markers.
10. Device for carrying out the process according to at least one of the preceding claims characterized in that it comprises a sample library including at least one enclosure in which the samples collected in step (a) of the process are stored and preserved, in inert containers each closed by a stopper and in that this enclosure is suitable for placing these samples under given temperature, pressure, humidity, atmosphere conditions.
