Page 143 - Intelligent Digital Oil And Gas Fields

106 Intelligent Digital Oil and Gas Fields

Table 4.3 Summary of AIPA Families and Techniques.
Family Specific Technique

Computational intelligence Neural networks
Fuzzy systems
Evolutionary computation
Data mining
Rule-based case reasoning Bayesian networks
Expert systems
Automatic process control Classical
Robust
Adaptive
Intelligent
Stochastic
Workflow automation
Proxy models Surrogate models
Top-down models
Virtual environments
From Bravo, C., Saputelli, L., Rivas, F., P erez, A.G., Nikolaou, M., Zangl, G., et al., 2014. State of the
Art of Artificial Intelligence and Predictive Analytics in the E&P Industry: A Technology Survey. SPE
150314-PA, https://doi.org/10.2118/150314-PA.

In 2009, the Society of Petroleum Engineers (SPE) (the E&P flagship
professional organization) have established the AIPA subcommittee, within
its Digital Energy Technical Section, with the mission of promoting the
development and application of AIPA techniques in the oil and gas industry.
With increasing interest and uptake of AIPA technologies in oil and gas, in
2011, the subcommittee was promoted to a new technical section, named
Petroleum Data-Driven Analytics (PD2A). Bravo et al. (2014) have con-
ducted a comprehensive technology survey that provides the state of the
art of AIPA use in the oil and gas industry.
According to approximately 75% of respondents, management of large
volumes of data remains a major challenge of the E&P industry, mostly
because of the lack of integration in IT management and analysis. While
automated process control is perceived as the most productive and mature
AIPA technology in DOF programs worldwide, Fig. 4.1 indicates that data
mining, neural networks, workflow automation, fuzzy logic, and expert sys-
tems are the most recognized AIPA applications.
In particular, data mining appears to be the most familiar AIPA technol-
ogy, mostly in areas of data management and integration, data filtering,
cleansing and imputation, and information search.