Portable analysis system

Patent metadata

Field	Value
Publication number	US-12548640-B2
Application number	US-202217991960-A
Country	US
Kind code	B2
Filing date	Nov 22, 2022
Priority date	Nov 26, 2021
Publication date	Feb 10, 2026
Grant date	Feb 10, 2026

Abstract

The present invention refers to a portable analysis system that, from analytical methodologies set based on digital images acquired by portable devices and accessories developed in an expedited way, allows the performance of the analytical test without technical deepening by the executor of the analysis. Such methodologies allow the evaluation of chemical species relevant to the O&G sector, for following up compositional parameters in aqueous samples and analytical monitoring of production and effluent treatment. Thus, a quick implementation is possible considering eventual diagnoses associated with the parameters of interest and following up of different processes. The application can be in any laboratory present in an operational unit, whether offshore or onshore, as well as it can be used for direct measurement in the field, depending on the necessary environmental and operating conditions.

First claim

The invention claimed is:

1 . A portable analysis system comprising:
a chamber configured to:
hold a reaction mixture between a sample and a reagent solution inside the chamber; and
receive artificial light and avoid contamination from ambient light; and
a camera outside of the chamber and configured to capture digital images of the reaction mixture inside the chamber, wherein the system is configured to:
analyze the digital images of the reaction mixture for a reaction mixture property, wherein the reaction mixture property comprises color, turbidity, or combinations thereof; and
determine a sample property using the analysis of the digital images, wherein the sample property comprises alkalinity, analyte concentration, or combinations thereof.

2 . The system according to claim 1 , wherein analyzing the digital images of the reaction mixture comprises using regression algorithms developed with computer vision and machine learning techniques for determination of the sample property.

3 . The system according to claim 2 , wherein the regression algorithms are developed with deep artificial neural networks, wherein the deep artificial neural networks are trained by supervised reverse propagation.

4 . The system according to claim 1 , wherein the chamber is physically created by an additive manufacturing process.

5 . The system according to claim 1 , wherein analyzing the digital images of the reaction mixture comprises using predictive models trained from calibration data.

6 . The system of claim 5 , wherein the predictive models are adjusted using the captured digital images.

7 . The system of claim 1 , further comprising a mobile device comprising the camera.

8 . The system according to claim 7 , wherein the mobile device is configured to instruct the user and/or analyst and assist the user and/or analyst in capturing the digital images.

9 . The system of claim 7 , wherein the mobile device is a cell phone.

10 . The system of claim 9 , further comprising a cell phone support configured to support the cell phone in a position with respect to the chamber such that the camera of the cell phone is configured to capture digital images of the chemical reaction inside the chamber.

11 . The system of claim 1 , wherein the analyte concentration comprises phosphate concentration, chloride concentration, sulfate concentration, or combinations thereof.

Assignees

• Petroleo Brasileiro Sa Petrobras
• Univ Federal Do Rio De Janeiro Ufrj
• Petroleo Brasileiro S A —Petrobras
• Univ Federal Do Rio De Janeiro—Ufrj

Inventors

• Mesquita De Carvalho Rogerio
• Alves Fontes Rosane
• Vasques Peixoto Vieira Helida
• Dos Santos Evangelista Raphael
• Aleixo Gonçalves Maria Luisa
• Magliano Balbi De Faria Rafaella
• Da Gama Leitão Helena Cristina
• Augusto Frata Fernandes Leandro
• Suarez Ferreira Stefan
• Machado De Azevedo Novaes Aline
• Gomes Teixeira Marco Antônio
• Domingos Rodrigues Igor
• Lelis Da Silva Gustavo
• Dos Santos Silva Caroline

Classifications

• G07C3/08

  Registering or indicating the production of the machine either with or without registering working or idle time · CPC title

• G06N5/04

  Inference or reasoning models · CPC title

• G06N3/084

  Backpropagation, e.g. using gradient descent · CPC title

• G06N3/09

  Supervised learning · CPC title

• G16C20/70 Primary

  Machine learning, data mining or chemometrics · CPC title