EditorialAutonomous agents and computational intelligence: the future of AI application for petroleum industry
Section snippets
Aim and scope
The special issue is dedicated to artificial intelligence (AI) applications in the petroleum industry. The papers in this issue are extended versions of selected papers presented at the First Workshop Intelligent Computing in the Petroleum Industry held in Mexico-city, November 24–25, 2002. About 20 papers were presented at the Workshop by the authors from 7 countries, from which 10 papers were carefully selected by the International Program Committee for this special issue. The primary goal of
Oil and gas exploration
Oil exploration is a large-scale human activity in which the acquisition, circulation and use of knowledge are more critical for decision-making. With oil and gas companies presently recovering, on the average, less than a third of the oil in proven reservoirs, any means of improving yield effectively increases the world's energy reserves. Accurate reservoir characterization through data integration (such as seismic and well logs) is a key step in reservoir modeling and management and
Process industry
Multi-agent architectures have been reported for applications where distributed decision-making is advantageous such as flexible manufacturing control, planning and scheduling, diagnosis, process design, modeling and diagnosis, and supply chain management (Braunschweig & Gani, 2002). To date, most of these systems are exploratory in nature. Multi-Agent Systems (MAS) are specially adequate for the solution of problems with a dynamic, uncertain and distributed nature. Within industrial
Evacuation in risk zone
A lot of different kinds of potentially dangerous situations permanently emerge in different regions and countries. Oil companies are also exposed to this type of situations. For example, the zone of the gulf of Campeche, the principal oil reservoir in Mexico, is characterized by a high degree of possibility of meteorological cataclysms (like hurricanes). For over a decade, information systems for computer support of decision-making in such type of situations have been developed for planning,
Knowledge management within the oil company
The competitiveness of large companies and organizations heavily depends on how they maintain and access their knowledge (i.e. their corporate memory). Most information in modern electronic media is based on text, audio and video and rather weakly structured. Finding and maintaining information is a hard problem in this weakly structured representation media. Moreover, corporative knowledge and experiences should be integrated also from individual knowledge and experiences of the employer. Raw
Enabling technologies: soft computing and pervasive computing
Finally, the works on Soft Computing and Pervasive Computing belong to the technologies, which can be widely used in different application domains within the petroleum sector.
The work of Ildar Batyrshin, on Linguistic Representation of Quantitative Dependencies is an important example of the application of soft computing techniques, particularly a novel type of fuzzy rules like “If X is SMALL then Y is QUICKLY INCREASING”. These rules are based on a granulation of directions of function change
Future trends: agent-based computing and soft computing
During the round table discussion on the future of the AI technologies in the petroleum industry held within the Workshop, the consensus was achieved that Agent-Based Computing, Soft Computing and hybrid technologies are the main candidates to deal with the main challenges in the field. Here we stress some trends grounded by the works presented in this issue.
The last 25 years we have seen the appearance of several paradigms to design software systems such as Procedural Programming, Structured
References (3)
Cited by (12)
Technical research on realizing remote intelligent diagnosis of petroleum drilling loss circulation under smart city strategy
2021, Future Generation Computer SystemsCitation Excerpt :The model based on the three-level well control theory, which can be used as an aid to prevent future blowouts and assess the risk related to these barriers [17]. In recent years, hybrid intelligence systems that integrate different Ai technologies have become increasingly accepted in the mainstream of the petroleum industry due to their ability to deal with real-world complexities involving imprecision, uncertainty, and ambiguity [18–21]. The oil drilling operation is a complicated system, which is often accompanied by fuzziness and randomness [22,23].
Monitoring and control strategies to manage pressure fluctuations during oil well drilling
2018, Journal of Petroleum Science and EngineeringCitation Excerpt :Thus, control and automation of drilling operations are necessary for the future challenges of petroleum engineering, especially under a scenario of narrow operating windows. An analysis of the literature reveals that most of the papers employ monitoring (Alvarado et al, 2004; Mohaghegh, 2005; Nikravesh et al, 2002; Zhang et al., 2004; Sheremetov et al, 2005 e Sheremetov et al, 2008, Hermann, 2014), without coupling with diagnosis analysis and decision making (control) studies. The present paper presents the efforts aiming oil well drilling automation.
Modeling of the charging characteristic of linear-type superconducting power supply using granular-based radial basis function neural networks
2012, Expert Systems with ApplicationsCitation Excerpt :Computational intelligence (CI) has emerged as a consortium of highly visible and dominant information technologies of granular computing, neuro-fuzzy computing and evolutionary and biologically inspired computing. CI addresses both the conceptual challenges and the computing needs arising within the realm of software, applied engineering and information technology (Alvarado, Cheremetov, & Cantú, 2004; Pedrycz & Vasilakos, 2000; Quah & Sriganesh, 2008; Santos, Coste, & Coelho, 2007; Scotti & Piuri, 2009; Thammano & Moolwong, 2010). Among the CI technology, radial basis function neural networks (RBFNNs) have gained much popularity due to their abilities to approximate complex nonlinear mappings while retaining a relatively simple and transparent topology of the networks (Balasubramanian, Palanivel, & Ramalingam, 2009; Dhanalakshmi, Palanivel, & Ramalingam, 2009; Guillén et al., 2010; Jung, Lim, & Kim, 2009; Kamalasadan & Ghandakly, 2007; Khajeh & Modarress, 2010; Lee, Chiang, Shih, & Tsai, 2009; Lin, Wang, Chen, Chen, & Yen, 2009; Mehrabi, 2009; Subashini, Ramalingam, & Palanivel, 2009; Weatherspoon & Langoni, 2008; Wu et al., 2010; Yilmaz & Özer, 2009; Zhu, Zhang, & Bao, 2000).
Fuzzy expert system for solving lost circulation problem
2008, Applied Soft Computing JournalTransient Open-Closed Loop Experimental Validation of a Nonlinear Two-Phase Flow Distributed System
2023, Macromolecular Reaction EngineeringSmart monitoring and decision making for regulating annulus bottom hole pressure while drilling oil wells
2016, Brazilian Journal of Chemical Engineering