Due to the emergence of new microbreweries in the Brazilian market, there is a need to construct equipment to quickly and accurately identify the alcohol content in beverages, together with a reduced marketing cost. Towards this purpose, the electronic noses prove to be the most suitable equipment for this situation. In this work, a prototype was developed to detect the concentration of ethanol in a high spectrum of beers presents in the market. It was used cheap and easy-to-acquire 13 gas sensors made with a metal oxide semiconductor (MOS). Samples with 15 predetermined alcohol contents were used for the training and construction of the models. For validation, seven different commercial beverages were used. The correlation (R2) of 0.888 for the MLR (RMSE = 0.45) and the error of 5.47% for the ELM (RMSE = 0.33) demonstrate that the equipment can be an effective tool for detecting the levels of alcohol contained in beverages.

No abstract

Indication of peach maturity and inspection of fruit quality in an orchard are generally analyzed based on the farmer's experience, which may be subject to failure and result in financial losses due to negligence or late harvest. Volatile organic compounds (VOCs) vary in quantity and type, depending on the different phases of fruit growth. Thus, the electronic noses are an alternative, since they allow the online monitoring of the VOCs generated by the culture. The correlate works found in the literature focus on the detection of post-harvest peach maturation, not the monitoring of fruits in the orchard, to detect the stage of growth and maturation, and that, when measuring these properties, methods are generally used destructive instruments. Analyzing this context, we developed a prototype for the identification of the maturation of the Eragil peach in the pre-harvest growth cycle. The prototype was built containing thirteen metal oxide semiconductor gas sensors. The pre-processing method for feature selection was applied, Pearson's Chi-square test, which provided the reduction for six sensors. Based on the Random Forest method with a Linear Discriminant Analysis, we reduced the dataset for six sensors and obtained a 1.92% error rate in the sample test step. It shows that the device could be optimized to this application and confirm that it is promising for local monitoring of fruit ground-based on the emission of VOCs and that several devices of this in-network can provide information for an optimized harvest.