THE FUTURE
OF THE OIL INDUSTRY
Oil Prospection
The application of the study of microfossils to the area of oil prospection had its first appearance in 1890 in Poland [21], but it was in the USA, in 1920, with the use of microfossils to identify the age of probes extracted from drilling rigs, that a bigger advance in the development of the field of Applied Micropaleontology was attained.
Wellsite biostratigraphy is applied in deciding the total depth of a well (to stop drilling), predicting the approach of a known overpressured horizon, determining casing points, determining the presence of faults , thrusts (repeated sections), correlating the drilled sections with nearby wells, or as a rapid means of providing a biostratigraphical age subdivision or palaeoenvironment for the drilled sequence.
Oil Rig Operation
In geological environments of marine sedimentation, the stratigraphic reference of greater accuracy is the systematic identification of planktonic foraminifera fossils for their cosmopolitan dispersion characteristic.
When drilling a well, the stratigraphic control with the identification of geological timelines is of great importance not only in pioneer drilling, but also in the process of developing a productive oil field.
The analysis with the objective of identifying biozones is anachronic, laborious, and time-consuming. It requires a team of experts that, nowadays, makes it impossible to apply directly to oil exploration. Drill cuttings or cores (increasingly rarely obtained) must be taken to a land laboratory for processing, separation, selection and classification of microfossil association.
Over time, as the exploratory activity has intensified, operating costs have dramatically increased, seismic processes have evolved and sophisticated. However, the stratigraphic control done by fossil records has been relegated to a second plan, because it has not followed the rhythm required by the modern oil industry, commanded by the motto "faster & cheaper".
The Challenges of the Oil Industry
Around the world billions of people are affected by one of the worst health crises of the past century. The global economy is under pressure in ways not seen since the Great Depression in the 1930s. Even assuming that travel restrictions are eased in the secondhalf of the year, we expect that global oil demand in 2020 will fall by 9.3 million barrels a day (mb/d) versus 2019, erasing almost a decade of growth.
Oil producers in the OPEC+ group agreed to cut output by an initial 9.7 mb/d versus their agreed baseline, effective 1 May 2020. If production does fall sharply, some oil goes into strategic stocks, and demand begins to recover, the second half of 2020 will see demandexceed supply.
Oil companies must expand online program that teaches its employees artificial intelligence skills, part of an effort to cut costs, improve business processes, and generate revenue. Artificial intelligence enables the company to process the vast quantity of data across the businesses to generate new insights, which can keep the ahead of the competition.
Oil corporations can potentially gain crucial insights to improve their business outcomes in their upstream processes with the integration of AI software. The AI tools can help oil and gas companies in automate the analysis of the gathered geological data. Organizations across the world are trying to make the exploration and the production processes more efficient and optimized. The operations in this field are the major factors that are driving the usage of Artificial Intelligence (AI) in oil and gas companies.
Deep learning, object recognition and micropaleontology
The 3D object recognition area has, in the last few years, experienced a growing boosted by the increased availability of new algorithms and models, 3D data and the popularization of a varied palette of 3D sensors.
Methods developed in this area find application in a wide range of areas, from the field of robotics to the security and surveillance domain.
The general tendency in this area has been the use of Deep Learning (DL) techniques. DL is a form of machine learning that enables computers to learn from experience and understand the world in terms of a hierarchy of concepts. DL employs very deep CNNs, with neural networks that sometimes consist of more than 100 layers, in contrast to the Artificial Neural Networks (ANNs).
Because DL CNNs gather knowledge from examples, there is no need for a human computer operator to formally specify all the knowledge that the computer needs. The capacity to represent a hierarchy of concepts in a network dozens of CLs deep allows a DL CNN to learn complicated concepts by building them out of simpler ones; a graph of these hierarchies would be many layers deep.
Based on 3D digital images, several analysis and simulations can be performed and, processes currently performed manually and exhaustively, can be automated and could be performed and validate a method for fully automated microfossil identification, providing a relevant tool to oil exploration operation.
An Unprecedented Design:
Automated Identification of Microfossils
“Automated Identification & Classification of microfossils in support of the Oil & Gas exploration.” Designed to a real time support to the offshore oil rig.
The objective of the project is to structure, develop, and pilot a microfossil automated identification & classification system, to provide Real Time operational support to the activities of exploration and production of oil & gas, in the offshore environment. The description of its fundamentals and future operations focuses on the identification and classification of planktonic foraminifera, however, if successful, it can also be employed to identify other biological species.
The system provides the study and architecture of deep learning neural network for identification & classification of biological markers, from digital images from real samples, through automated selection and classification of fossil species, indicators of geological time line (biozones), contained in 2D/3D digital images.
The project, with features of R&D, offers practical objectives and intend to be applied in the oil industry, based on solid foundations of biostratigraphy, digital image processing of robot operations, supported by deep learning neural network systems.
Finally, the unprecedented proposal is based on the knowledge integration of advanced fields of science (paleontology, graphic and visual computing, and Learning Machine). Therefore, its development will require trials in the mentioned fields to determine the most appropriate actions to accomplish such purposes.
Deep learning, object recognition and micropaleontology
The 3D object recognition area has, in the last few years, experienced a growing boosted by the increased availability of new algorithms and models, 3D data and the popularization of a varied palette of 3D sensors.
Methods developed in this area find application in a wide range of areas, from the field of robotics to the security and surveillance domain.
The general tendency in this area has been the use of Deep Learning (DL) techniques. DL is a form of machine learning that enables computers to learn from experience and understand the world in terms of a hierarchy of concepts. DL employs very deep CNNs, with neural networks that sometimes consist of more than 100 layers, in contrast to the Artificial Neural Networks (ANNs).
Because DL CNNs gather knowledge from examples, there is no need for a human computer operator to formally specify all the knowledge that the computer needs. The capacity to represent a hierarchy of concepts in a network dozens of CLs deep allows a DL CNN to learn complicated concepts by building them out of simpler ones; a graph of these hierarchies would be many layers deep.
Based on 3D digital images, several analysis and simulations can be performed and, processes currently performed manually and exhaustively, can be automated and could be performed and validate a method for fully automated microfossil identification, providing a relevant tool to oil exploration operation.
Looking for
Thanks for climbing to the final banner.
After several years negotiating with PETROBRAS the financing to complete the project, the pandemic turmoil seems to frustrate my expectations. Its completion requires the involvement of Deep Learning experts and investors who believe in its economic potential.
This home page aims to draw the attention of partners interested in finalizing the project “Automated Identification & Classification of microfossils in support of the Oil & Gas exploration.”
What I'm looking for:
-
Partners with expertise in Deep Learning Machine in identification and selection of 2D/3D digital images
-
Financing to performer the final stages of the project and to conduct the offshore pilot tests
What is offered:
-
An opportunity to generate a high sophisticated tool
-
Being the owner of a very profitable operating system
-
Solid relationship with PETROBRAS* (as a potential partner)
-
Full mastery of the technique and the system usage
-
Collection of Marker Species from Brazilian sedimentary basins
-
Full-time apply in the project
My best regards
PETROBRAS* - Petróleo Brasileiro SA. – Brazilian State owner oil corporation.
Prof. Juarez Fontana MSc. PhD.
-
Geologist & Micropaleontologist
-
Geology graduation – 1967 – RGS Federal University
-
MSc. Economic Geology - 1987– São Paulo University
-
PhD. Natural Resources Management – 1997 – Campinas University – SP
-
Oceanography – University of Whashington - USA
-
Former Head of Plank Foraminifera Department of PETROBRAS Research Center (CENPS)
-
Published the first Plank Foraminifera biostratigraphy scale of the Brazilian offshore sedimentary basins.
-
Graduation and PostGraduation Courses: professor in Geology and Oceanography courses - Parana Federal University; UNESP University; São Judas University.
-
Principal of Bachelor Course of Technology of Oil & Gas, Geoscience and Oceanography in São Judas University – SP.
About the Author
