Journal of King Saud University – Science 33 (2021) 101542

Contents lists available at ScienceDirect

Journal of King Saud University – Science

journal homepage: www.sciencedirect.com

Review

AI-based geo-engineering integration in unconventional oil and gas

Shiying Di a, Shiqing Cheng a,⇑, Nai Cao b, Chaoli Gao c, Linan Miao d
a
State Key Laboratory of Petroleum Resources and Prospecting, China University of Petroleum, Beijing 102249, China
b
Sinopec Research Institute of Petroleum Engineering, Beijing 100101, China
c
Research Center of Yanchang Oilfield Co., Ltd, Yan ’An, 717600 Shaanxi, China
d
Daqing Logging and Testing Service Company, Daqing 163458, China

a r t i c l e i n f o a b s t r a c t

Article history: In the oil and natural gas industry, artificial intelligence (AI) technology has penetrated into all links from
Received 7 October 2020 exploration and development to construction sites. This paper investigates the application of artificial
Revised 3 May 2021 intelligence in each link of oil reservoir. It is found in investigation that each algorithm plays a different
Accepted 6 July 2021
and important role at each stage and every link of oil accumulation development cannot leave the coop-
Available online 17 July 2021
eration of artificial intelligence. But, the application of AI is mostly scattered, forming a physical isolation,
a lot of single information-island. This situation increases the communication cost of cross-departmental
Keywords:
data cooperation and the repetitive screening and recognition work has seriously affected work effi-
Artificial intelligence algorithm
Unconventional reservoir
ciency. To address unsolved problems with the current application of AI, AI-based geo-engineering inte-
Engineering integration gration in unconventional oil and gas are proposed this article considers. Integrate data islands, and
Geo-engineering integration realize internal resource sharing, treat exploration and development as an organic whole, extend explo-
Exploration and development ration to development. This article takes the well factory operation mode, meanwhile, the real-time syn-
chronization and coordination of all links has been fully realized. This kind of integration of geology and
engineering is helpful to realize coordination and cooperation at all levels, regions, and disciplines, effec-
tively benefiting development of unconventional oil and gas reservoirs.
Ó 2021 The Author(s). Published by Elsevier B.V. on behalf of King Saud University. This is an open access
article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
2. The trend of geo-engineering integration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
3. A powerful tool-AI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
3.1. Why supportive? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
3.2. Summary of commonly used AI technologies in the oil and gas sector. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
3.3. The influences of AI on the petroleum industry and oil and gas economic design. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
4. AI and geo-engineering integration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
4.1. Application of AI in reservoir geology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
4.2. AI applications in engineering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
4.3. Application case analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
4.3.1. Western South China Sea Oilfield . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
4.3.2. Dagang complex fault block oil field . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
4.3.3. Marine shale gas in southern China (Fuling shale gas) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
4.3.4. Green field heavy oil asset in Kuwait . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

⇑ Corresponding author at: 18 Fuxue Road, Changping District, Beijing, China

Peer review under responsibility of King Saud University.

Production and hosting by Elsevier

https://doi.org/10.1016/j.jksus.2021.101542
1018-3647/Ó 2021 The Author(s). Published by Elsevier B.V. on behalf of King Saud University.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
S. Di, S. Cheng, N. Cao et al. Journal of King Saud University – Science 33 (2021) 101542

5. Conclusions. . . . . . . . . . . . ....... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Funding . . . . . . . . . . . . . . . . ....... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Declaration of Competing Interest . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
References . . . . . . . . . . . . ....... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8

1. Introduction deployment, integration of exploration and development manage-

ment, integration of exploration and development research, inte-
AI began in the 1960s and enjoyed a notable success. AI pro- gration of reserves and production. In the specific work, the
vides new ideas to unconventional oil and gas development (Li integration of production and scientific research, the integration
et al., 2020; Yang and Zou, 2019). Liu et al. (2018) use hybrid intel- of engineering and geology, and the integration of oil testing with
ligent systems (HIS) to optimize the development effect is the main development” is the overall framework of the main content (Li
application method of AI in the field of reservoir development and et al., 2020; Redutskiy, 2017; Wan et al., 2020; Yao, 2014).
development. AI also provides a more accurate method for select-
ing the level of stimulation measures. Xinjiang Oilfield Company
adopted BP-NN and LM algorithms, and used GR to optimize the 3. A powerful tool-AI
construction plan, and realized the scientific decision-making and
fine optimization of the parameters of the fracturing plan (Maity Artificial intelligence (AI) refers to intelligently developed
and Ciezobka, 2019). Al-Fattah developed a rigorous and advanced machines that is programed to learn, think, perceive, and solve
data-driven model called GANATS which can describe and predict problems like humans. With the leap in computer processing capa-
global crude oil demand. Al-Fattah (2021). After oil and gas reser- bilities and the continuous innovation of algorithms, AI technology
voirs are exploited, crude oil transportation and preservation are has achieved unprecedented development and penetrated into all
key issues. An intelligent simulation decision-making system for aspects of life. In the energy sector, AI is indispensable as well.
crude oil storage and transportation scheduling is designed based
on the C++ Builder 6 program development platform (Yuan,
2018). Ahmadi et al. (2012) developed a unified particle swarm 3.1. Why supportive?
optimization method based on neural networks to predict asphal-
tene precipitation, which method is based on ANN to predict Artificial intelligence (AI) refers to intelligently developed
asphaltene qualitative precipitation. AI plays an important role in machines that is programed to learn, think, perceive, and solve
oil and gas storage applications (Koroteev and Tekic, 2021; problems like humans. AI algorithms are often used in oil reser-
Kuang et al., 2021; Maryam and Abbas, 2018). voirs. Each algorithm has its own principle.
However, the above methods each form a system, there is cur- ANN is the ability to model complex non-linear processes with-
rently no systematic summary of AI in the integration of geological out the need to establish any relationship between input and out-
engineering in unconventional oil and gas (Xie et al., 2017; Chen put variables. Fuzzy Logic can use ‘‘human language” to describe
et al., 2021). This study analyzes AI in the construction of reservoir problems and their fuzzy solutions. GR is simple chromosome-
geo-engineering and the cases of AI-based integrated work plat- like data structures that encode potential solutions to specific
forms. Finally, the future research directions and suggestions for problems. The principle of SVM is based on statistical learning the-
the integration of artificial intelligence in unconventional oil and ory and structural minimization. Case-based reasoning (CBR) is to
gas geological engineering are put forward. It provides a reference reuse one or more similar cases that have been previously solved in
for the application of artificial intelligence integration in uncon- new problems.
ventional oil reservoirs. The efficient application of AI requires combined use of algo-
rithms for optimization purposes. Meanwhile each question is
unique, and model developed needs to consider the nature of the
2. The trend of geo-engineering integration problem.

The status of unconventional oil and gas in global energy supply

has become increasingly prominent and has become an important 3.2. Summary of commonly used AI technologies in the oil and gas
part of oil and gas production, reshaping the global energy map sector
and pattern (Jia, 2017; Tong et al., 2018; Zou et al., 2018).
Unconventional oil and gas, taking tight oil with higher heat as In order to understand the characteristics of each algorithm in
an example, strong heterogeneity, poor seepage ability, low single depth and provide a basis for algorithm optimization in AI oilfield
well production, low oil and gas field recovery rate and unstable exploration and development, the characteristics of some widely
production are also among its characteristics (Jiao, 2019; Yang used algorithms in the field are summarized. The application of
and Zou, 2019; Sun, 2019). In the traditional oilfield management artificial intelligence in the reservoir can be classified as follows
model, the continuous process of geology and engineering is artifi- according to the algorithm.
cially divided, and there is a lack of inheritance and reciprocating ANN are used for reservoir characterization, optimization
correction process between the data. If the construction result of design of production stimulation measures, drill bit selection,
the back-end project cannot be fed back to the geology department and optimization of field operations. Fuzzy logic has been used
in time, the reservoir model cannot be revised in time (Yang et al., reservoir characterization, production enhancement measures,
2020). enhanced oil recovery, and decision analysis. Genetic algorithms
Taking China as an example, all major oil fields throughout is used for real-time optimization of the entire oil and gas produc-
China seeks to become new kinds of oil fields under the integrated tion process. Support vector machine methods commonly used in
management model of geo-engineering. It successfully established oilfields are used to identify flooded layers. The literature shows
and realized the ‘‘ integration of exploration and development that drilling plans can be optimized and executed through CBR.
2
S. Di, S. Cheng, N. Cao et al. Journal of King Saud University – Science 33 (2021) 101542

However, it is worth noting that in practical applications, AI 4. AI and geo-engineering integration

may have defects such as difficulty in reaching the optimal solu-
tion, overfitting, and large demand for computing capacity. In the concept of unconventional oil and gas reservoir geologi-
cal engineering integration, ‘‘geology” and ‘‘engineering” are not
synonymous with disciplines in a narrow sense. ‘‘Geology” broadly
3.3. The influences of AI on the petroleum industry and oil and gas refers to the comprehensive multi-disciplinary research work cen-
economic design tered on reservoirs, including reservoir characterization, geological
modeling, geomechanics, and reservoir engineering evaluation.
The influences of AI on the economic design of the oil and gas ‘‘Engineering” refers to a series of engineering technologies from
industry is reflected by the selection of one or more models for drilling to production in the process of exploration and develop-
comparison and correction through value assessment. For exam- ment, including drilling, fracturing, oil production, reservoirs, eco-
ple, what accounts for the increase in costs? Through the analysis nomic evaluation, etc. Under the traditional oilfield management
of the single well data with the big data of artificial intelligence in model, geological and engineering data are physically and logically
each oil field, it can be understood that the increase of relevant separated. Having separately stored data, data are analyzed inde-
costs is directly related to the actual production. At the same time, pendently by different departments. It results in low communica-
the artificial intelligence system is used to simulate the impact of tion efficiency and duplication of work (Fu et al., 2019; Li et al.,
each oil field’s production capacity construction on the ROI, and 2019; Shu et al., 2020).
finally the reasonable economic design is obtained. The integration of geo-engineering (Fig. 1) requires interaction
The design of oil well and petroleum industry by artificial intel- between geology and engineering, which provides a basis for engi-
ligence is mainly embodied in the overall optimization of new and neering operations such as drilling and fracturing. As the progress
old data. Through the value evaluation, the benefit index of each of engineering operations changes, timely updating of new data
scheme is determined to provide scientific basis for the final opti- also helps to update the geological model in real time. Because
mization decision. According to the specific situation, one or more each link of the exploration and development of unconventional
modes can be selected for mutual comparison and correction. oil and gas fields involves a large amount of heterogeneous data,

Fig. 1. The integration of gen-engineering.

Table 1
Application of AI in reservoir geology.

Geological link method Author Application

Lithology identification CNN (Chen et al., 2020; Wu et al., 2021) Characterization of tight reservoirs, low resistivity
INN Kulga et al. (2018) pay identification
FSVM Cao et al. (2018)
SVM + ANN Han et al. (2017)
SVM + BP Ge et al. (2019)
Reservoir evaluation SVM + ANN + RBF HIS Duan et al. (2020) Reservoir physical parameters Reservoir physical
PSO + SVM Yin (2016) parameters
Hybridgenetic algorithm Yin et al. (2017)
Crack identification A NN (Ahmed et al., 2019; Jing et al., 2020; Crack pressure, closure pressure
Nanda et al., 2019)
ANN + Fuzzy Fan (2016) Three-dimensional crack distribution
CNN Wu et al. (2018) Fractured reservoir category
ML + DBA + S VM Anderson et al. (2016) Automatically identify cracks
BP Huang et al. (2016) Crack density

3
S. Di, S. Cheng, N. Cao et al. Journal of King Saud University – Science 33 (2021) 101542

Fig. 2. Learners integrated with different intelligent algorithms.

Fig. 3. Geological analysis data intelligent platform.

the massive data processing advantage of AI brings opportunities soning algorithms to achieve better application effects. There are
for real-time communication and collaborative processing of all still deficiencies in geological evaluation based on artificial intelli-
links of geological engineering. gence, and the classification accuracy and scope of application
need to be further improved. Algorithm optimization and opti-
4.1. Application of AI in reservoir geology mization are still the direction of AI research. At the same time, this
selection and integration approach is applicable to all aspects of
Reservoir-centric geological modeling and reservoir characteri- geological engineering and has great application prospects.
zation are the basis for engineering construction. With the techno- In the research of reservoir geology, AI is not only applied to a
logical progress of exploration and development, the foundation of single technical link, but also gradually applied to the construction
geological data, application scenarios and application purposes are of a geological intelligence platform, which partially reflects the
constantly undergoing profound changes. Table 1 summarizes part concept of integration. The application of AI in logging data inter-
of the AI technology application research involved in reservoir pretation and other links is the technical preparation and founda-
geology. tion for the ultimate realization of geo-engineering integration.
Generally speaking, artificial neural network algorithm is one of In order to enable the sharing of reliable geological information
the most widely used algorithms in the geological link of uncon- obtained by AI in the geological process, and to improve the infor-
ventional oil reservoirs. The research ideas of ensemble learning, mation management level and data processing efficiency of oil-
especially selective ensemble learning, are gradually used by fields, a geological analysis data intelligent platform came into
researchers (Li et al., 2021). At present, learners integrated with being as the technical foundation of geo-engineering integration
different intelligent algorithms have been used to classify lithol- (Zhao et al., 2018) (Fig. 3).
ogy. Learners based on lithology classification are used to further The platform is a basic preparation for the integration of geol-
identify the physical properties of the reservoir and combine the ogy and engineering. Its idea is to use artificial neural networks
geological information to determine the weight of each intelligent (ANN), BP neural network methods, support vector methods,
method in the learner (Fig. 2). genetic algorithms and other artificial intelligence methods to dis-
It is evident that scholars nowadays are no longer limited to the play seismic interpretation results, structural models, and three-
application of traditional methods, but are constantly trying new dimensional visualization of attribute models, dynamic models,
methods, seeking more general knowledge representation and rea- dynamic production profiles, etc. Moreover, the platform can
4
S. Di, S. Cheng, N. Cao et al. Journal of King Saud University – Science 33 (2021) 101542

Table 2
Application of AI in engineering.

Engineering link method application

Drilling ANN (Gidh et al., 2012) Drilling plan optimization, drill bit optimization, wellbore stability, construction
optimization
BP + GA (Opeyemi, 2016) Drilling platform selection, stuck control and corrective measures
CBR (Tiago et al., 2012) Real-time risk monitoring and decision-making
Wellpattern design QGA Layout and optimization of well patterns
ANN (Saputelli et al., 2002)
PSO (Mellit et al., 2009) Preferred measures
GA Well spacing involves
fracture BP Optimal fracturing parameters,
BP + GA Fracturing effect prediction
ANN (Saputelli et al., 2002; Wei et al., 2020) Fracturing effect
test CNN (Al-Kaabi et al., 1993; Juniardi et al., Wellbore dynamic liquid level recognition
1993)
Capacity evaluation ANN BP(Mohaghegh et al., 1994) Water saturation, flow law, reservoir dynamic analysis capacity forecast
FCM + SVM (Anifowose et al., 2011; Shi et al., Reservoir dynamic analysis (water content)
2018)
GNN + IPSO (Ahmadi, 2012) Capacity forecast
SVM (Ahmadi et al., 2013; Ahmadi, 2015)
ANN + FCM (Sebakhy and Emad, 2009)
PCA + APSO + LSSVM (Ahmadi, 2015) Production management

correlate the change rules between data, predict development indi- 4.3. Application case analysis
cators, evaluate the effects of measures, adjust process parameters,
tap the potential of reservoir production, achieving the goal of Although the idea of integration has been reflected in the two
data-driven decision-making in oil fields. links of geology and engineering, the true realization of AI geo-
engineering integration still requires further interaction and inte-
4.2. AI applications in engineering gration of the two links. Currently, practitioners around the world
are constantly conducting integrated exploration of unconventional
Due to many production problems in the development of oil oil and gas. In Sichuan China, Fuling marine shale gas, Dagang Oil-
fields, engineering technology optimization is imminent. Engineer- field and other complex oilfields have combined their own charac-
ing optimization can improve technology, increase capital returns teristics and made continuous efforts to integrate development
and ensure the stability of manpower. With the help of artificial compatible with their own geological and production characteris-
intelligence analysis and optimization engineering theory and tics. For the integration of geo-engineering in unconventional oil
method, to promote the intelligent transformation and upgrading and gas, further promotion of chemistry has certain practical signif-
of petroleum industry. icance. In the meantime, Green field Heavy Oil Asset in Kuwait ’s
The ‘‘ engineering ” link is to screen, optimize and guide the data integration management ideas in the oilfield planning and
implementation of a series of drilling and production, as well as construction process have seen more gratifying applications. Below,
to assess solutions. It makes continuous adjustment and improve- the application cases are analyzed to gain experience and explore
ment of engineering technical solutions, enriching knowledge and future research directions (Redutskiy, 2017; Wen et al., 2018; Liu,
engineering operation from drilling to fracturing in a systematic 2017; Ling et al., 2010; Salaheldin, 2018).
way. This paper focuses on the application areas of different AI
algorithms in drilling, fracturing, and productivity evaluation
4.3.1. Western South China Sea Oilfield
(Table 2).
The integration of AI has achieved remarkable application effect
Among them, drilling is an important construction link of the
in oil and gas production (Shen, 2018; Zhang and Zhu, 2013). Based
project, and its construction cost is high. The selection of equip-
on Western South China Sea Oilfield, the research and develop-
ment and plans before drilling, and the real-time monitoring of
ment of integrated functions such as test design, real-time opti-
construction during drilling will affect the drilling effect. A large
mization and adjustment, post-test productivity analysis and
number of factors that need to be considered in the prediction of
well test interpretation are carried out.
each link. Complex situations, and unpredictable construction con-
Conventional methods are not suitable for complex reservoirs
ditions all set barriers for drilling. AI has been successfully applied
in Western South China Sea Oilfield, the opening/closing time
in various aspects of drilling design, construction, and testing,
and production system allocation are unreasonable, and there is
which has improved the adaptability of the drilling process to com-
no combination of pre-drilling prediction and post-drilling correc-
plex situations.
tion. In response to the above problems, the study proposed the
Fracturing is the main measure and means to increase produc-
following four steps.
tion of oil and gas reservoirs, and high-efficiency fracturing is of
great significance for improving the recovery of unconventional
(1) Establish the prediction model of reservoir, fluid physical
oil and gas. To improve the effect and efficiency of fracture predic-
property and productivity parameters.
tion and optimize fracturing plan, it is necessary to introducing AI
(2) Study on test design method based on parameter prediction.
and machine learning methods to analyze the geological and engi-
(3) Research on real-time optimization and adjustment method
neering process parameters that affect the fracturing effect.
of test working system.
In the engineering process, it is also necessary to establish a col-
(4) Development of well test design and real-time optimization
laborative engineering decision-making and construction platform
decision system.
to realize the integrated technology construction of fracturing, dril-
ling and productivity evaluation, in order to achieve integration Based on the above 4 steps, the test experience of domestic and
and factorization, combined with the summary in Table 2. foreign oil fields is used for reference by this research. Big data
5
S. Di, S. Cheng, N. Cao et al. Journal of King Saud University – Science 33 (2021) 101542

ideas were applied, including reservoir and fluid property parame- struction of integration of geology and engineering and enable
ter prediction modules were established. The test plan is designed integration from concept to reality.
and the module is adjusted in real time. The productivity analysis In terms of AI application in each link, ANN and fuzzy logic are
and well test interpretation module and the knowledge generation applied to reservoir characterization. The quantitative characteri-
and management module are also included in the R&D software zation accuracy reaches 0.2 m. The application of fuzzy logic com-
system. Use the mature rules of the developed blocks to predict prehensively considers the development and distribution of
the physical properties of reservoirs and fluids, and design a rea- interlayers, the degree of reservoir production, the degree of water
sonable testing system (Fig. 4). According to the actual measure- flooding, the adaptability of the split injection process, and the
ment dynamics on site, real-time adjustment of the work system, needs of tertiary production. The method is applied to the
production capacity forecast analysis. decision-making analysis. In the engineering application, the sup-
This research can flexibly generate the parameter prediction port vector method is used to diagnose the working condition of
model, construct the knowledge base for parameter prediction, the pumping unit. The drilling is optimized and executed through
design the test scheme, determine the reasonable test process, col- CBR Plan and PSO develop new wells and drilling simulators, com-
lect the real-time data during the test and optimize the test work- bining parameter integration schemes to optimize on-site develop-
ing system. ment. ANN is used to select drill bits and optimize on-site
operations.

4.3.2. Dagang complex fault block oil field

Dagang Oilfield is a typical complex fault block oilfield. Geo- 4.3.3. Marine shale gas in southern China (Fuling shale gas)
engineering integration in Dagang Oilfield (Zhao et al., 2018) With long-term unremitting efforts, the Fuling shale gas field
(Fig. 5), the degree of waterflooding control increased from 63.5% has become the world’s largest shale gas field outside of North
to 80.1%, the degree of injection-production connectivity increased America. Through technological innovation, integrated innovation
from52.2% to 81.7%, and the injection-production response rate and management innovation, the integrated development model
increased from 65.5% to 80.2%. The labor productivity has is actively implemented. The entire working process and ideas
increased by 30%, and the system saves energy and reduces con- are based on exploration and development, surface and under-
sumption by more than 10%,provide practical guidance for the con- ground, scientific research and production integrated research

Fig. 4. Geo-engineering integration in Western South China Sea Oilfield.

Fig. 5. Geo-engineering integration in Dagang Oilfield.

6
S. Di, S. Cheng, N. Cao et al. Journal of King Saud University – Science 33 (2021) 101542

Fig. 6. Geo-engineering integration in Fuling Shale Gas.

and on-site implementation, achieving high-level, high-speed, ventional oil and gas geo-engineering integration can be combined
high-quality, and high-efficiency development (Xiao et al., 2018) with this phenomenon to explore the integration of old well areas,
(Fig. 6). old data and new data, and new technologies, ultimately achieving
In combine with the concept of geo-engineering integration, the a more comprehensive real-time shared database and algorithm
intelligent geological engineering platform is used to identify library.
address before fracturing, and the fracturing is performed in com-
bination with different geological conditions. Fracture construction
5. Conclusions
design, refined zoning and refined design are combined with each
other. During the fracturing construction, through the joint display
The application cases suggest the realization of geo-engineering
of the real-time occurrence of micro-seismic events and the
integration to some extent. However, the challenges faced still
dynamic changes of the construction curve, the construction situ-
require further assistance from high-tech such as AI. Based on
ation, the formation situation and the crack characteristics are ana-
summary work and case analysis, findings are summarized as
lyzed and judged. The next construction measures are adjusted to
follows:
ensure the construction Safe and smooth. Geological model is
updated via the implementation of monitoring fracturing effect
(1) Taking the well factory operation mode as an example, it is
and real-time adjustment and post-fracturing evaluation and anal-
of certain significance when combining historical successful
ysis (applying PSO and CBR). On this basis, use virtual well technol-
cases, establishing a standardized case model, forming a ref-
ogy to perform drilling and layout optimization on subsequent
erence sample library, and combining CBR and other tech-
development wells, and evaluate potential geological engineering
nologies for program design.
challenges and risks.
(2) The comprehensive algorithm has higher accuracy and bet-
The integrated application of geo-engineering in Fuling shale
ter effect than a single algorithm. The establishment of cor-
gas fully integrates engineering and geology, providing reference
responding algorithm libraries for each link and the
for other unconventional oil and gas resource development in
realization of different algorithm optimization for different
China.
reservoir structures are one of the potential directions of
geological engineering algorithm research.
4.3.4. Green field heavy oil asset in Kuwait (3) The unification of the formats of old data and new data, geo-
Green field is an attempt by young oil fields to integrate geo- logical data and engineering data is still one of the chal-
engineering data management. Green field Heavy Oil Asset in lenges faced in effective database construction. The
Kuwait uses data integration and visualization technology for Well, realization of data transformation and big data integration
Reservoir and Facility Management (WRFM) (Kharghoria et al., is the key to complete geo-engineering integration.
2019). It is still in the early stage of construction. The data volume (4) The previous artificial intelligence is very effective in every
is relatively small in comparison to mature and old oil fields, but aspect, but when each sector is applied to each other,
data integration in the practice of this oil field shows a clearer because each department is independent, the application
application prospect. or query of data will be delayed. The integration of geology
In accordance with the above three cases, it is clear that the data and engineering enables internal resource sharing, treating
mining and interaction potential of AI has not been fully utilized in exploration and development as an organic whole, which
terms of communication management and decision-making in the accelerates the implementation process of exploration and
construction of integration of geo-engineering. The integration is development, and shorten the conversion of resources to
mostly reflected in new wells, new blocks, and new oil fields. Even reserves, reduces the time for reserves to production conver-
in old oil fields such as Dagang Oil Field, production is increased sion, and saves funds spent on exploration and development
through the construction of new areas. The improvement and and oilfield construction.
application of old well and old data remains problematic. Further-
more, there is lack of insightful and deep analysis on the impact of Funding
the real-time update of engineering data on geological data.
Although there is work involving the matching of production The authors are grateful for financial support from the National
parameters and geological parameters, the research on the rela- Natural Science Foundation of China (11872073) and the National
tionships and the corresponding decision-making mechanism Science and Technology Major Projects of China (Grant Nos.
needs to be further strengthened. Therefore, in the work of uncon- 2017ZX05009-003).
7
S. Di, S. Cheng, N. Cao et al. Journal of King Saud University – Science 33 (2021) 101542

Declaration of Competing Interest Kuang, L.X., Liu, H., Ren, Y.L., et al., 2021. Application and development trend of
artificial intelligence in petroleum exploration and development. Pet. Explor.
Dev. 48 (1), 1–14.
The authors declare that they have no known competing finan- Kulga, B., Artun, E., Ertekin, T., 2018. Characterization of tight-gas sand reservoirs
cial interests or personal relationships that could have appeared from horizontal-well performance data using an inverse neural network. J. Nat.
Gas Sci. Eng. 59, 35–46.
to influence the work reported in this paper.
Li, G.X., Wang, F., Pi, X.J., et al., 2019. Optimized application of geology-engineering
integration data of unconventional oil and gas reservoirs. China Petrol. Explor.
24, 147–152.
Li, J., Zhou, Q., Williams, H., Xu, H., 2020. Back-to-back competitive learning
References mechanism for fuzzy logic based supervisory control system of hybrid electric
vehicles. IEEE Trans. Ind. Electron. 67 (10), 8900–8909.
Ahmadi, M.A., 2012. Neural network based unified particle swarm optimization for Li, Y., Lian, P.Q., Xue, Z.J., 2020. Application status and prospect of big data and
prediction of asphaltene precipitation. Fluid Phase Equilib. 314, 46–51. artificial intelligence in oil and gas field development. J. China Univ. Petrol. (Nat.
Ahmadi, M.A., 2015. Connectionist approach estimates gaseoil relative simulation in Sci. Ed.) 44, 1–11.
petroleum reservoirs: application to reservoir simulation. Fuel 140, 429–439. Li, H. et al., 2021. Applications of artificial intelligence in oil and gas development.
Ahmadi, M.A., Ebadi, M., Shokrollahi, A., Majidi, S.M.J., 2013. Evolving artificial Arch. Comput. Methods Eng. 28 (3), 937–949. https://doi.org/10.1007/s11831-
neural network and imperialist competitive algorithm for prediction oil flow 020-09402-8.
rate of the reservoir. Appl. Soft Comput. 13 (2), 1085–1098. Ling, L., He, W., Huang, W., 2010. Practices of highly efficient exploration in the
Ahmadi, M.A., Shadizadeh, S.R., 2012. New approach for prediction of asphaltene Kelameili gas field, Junggar Basin. Nat. Gas Ind. 30, 7–10.
precipitation due to natural depletion by using evolutionary algorithm concept. Liu, N.A., 2017. Application of geology-engineering integration in the development
Fuel. 102, 716–723. of tight gas reservoir in Sulige Gasfield. China Petrol. Explor. 22, 53–60.
Ahmed, A., Ali, A., Elkatatny, S., 2019. New artificial neural networks model for Liu, W., Yan, N., 2018. Application and influence of artificial intelligence in
predicting rate of penetration in deep shale formation. Sustainability-Basel 11, petroleum engineering area. Petroleum technology. 4, 32–40.
6527–6544. Maity, D., Ciezobka, J., 2019. Designing a robust proppant detection and
Al-Fattah, S.M., 2021. Application of the artificial intelligence gannats model in classification workflow using machine learning for subsurface fractured rock
forecasting crude oil demand for saudi arabia and china. J. Petrol. Sci. Eng. 200, samples post hydraulic fracturing operations. Journal of PetroleumScience and
108368. https://doi.org/10.1016/j.petrol.2021.108368. Engineering. 172, 588–606.
Al-Kaabi, A.U., Lee, W.J., 1993. Using Artificial Neural Networks To Identify the Well Maryam, S., Abbas, S., 2018. Evolving robust intelligent model based on group
Test Interpretation Model. In: Formation Evaluation, 26-28 September. Society method of data handling technique optimized by genetic algorithm to predict
of Petroleum Engineers, USA, pp, 233–240. asphaltene precipitation. J. Petrol. Sci. Eng. 171, 1211–1222.
Anderson, R.N., Xie, B., Wu, L., et al., 2016. Using machine learning to identify the Mellit, A., Kalogirou, S.A., Hontoria, L., Shaari, S., 2009. Artificial intelligence
highest wet gas producing mix of hydraulic fracture classes and technology techniques for sizing photovoltaic systems: a review. Renewable Sustainable
improvements in the Marcellus shale. In: Unconventional Resources Energy Rev. 13 (2), 406–419.
Technology Conference, 1-3 August. San Antonio, Texas, USA, pp. 3–11. Mohaghegh, S., Arefi, R., Ameri, S., et al., 1994. A methodological approach for
Anifowose, F.A., Ewenla, A.O., Eludiora, S.I., 2011. Prediction of Oil and Gas Reservoir reservoir heterogeneity characterization using artificial neural networks. In:
Properties using Support Vector Machines. International Petroleum Technology Annual Technical Conference and Exhibition, 25–28 September. Society of
Conference,15–17 November . https://doi.org/10.2523/IPTC-14514-MS. Petroleum Engineers, New Orleans, Louisiana, pp. 12–21.
Cao, Q., Wang, Z.Z., Fan, T.L., et al., 2018. Application of permeability interpretation Nanda, N., Kakkar, P., Nagpal, S., 2019. Computer-aided segmentation of liver
based on fuzzy support vector machine in sandy conglomerate reservoir. J. lesions in CT scans using cascaded convolutional neural networks and
Xinjiang Univ. (Nat. Sci. Ed.). 35, 264–271. genetically optimised classifier. Arab. J. Sci. Eng. 44 (4), 4049–4062.
Chen, G., Chen, M., Hong, G., Lu, Y., Zhou, B.o., Gao, Y., 2020. A new method of Opeyemi, Bello, 2016. Application of artificial intelligence techniques in drilling
lithology classification based on convolutional neural network algorithm by system design and operations: a state of the art review and future research
utilizing drilling string vibration data. Energies 13 (4), 888. https://doi.org/ pathways. In: Nigeria Annual International Conference and Exhibition, 2–4
10.3390/en13040888. August. Society of Petroleum Engineers, Lagos, Nigeria, pp. 23–41.
Chen, G., Wu, J., Liu, Y., et al., 2021. Geology-engineering integration key Redutskiy, Y., 2017. Integration of oilfield planning problems: infrastructure design,
technologies for ten billion cubic meters of shale gas productivity development planning and production scheduling. J. Petrol. Sci. Eng. 158, 585–
construction in the Southern Sichuan Basin. Nat. Gas. Ind. 41, 72–82. 602.
Koroteev, Dmitry, Tekic, Zeljko, 2021. Artificial intelligence in oil and gas upstream: Salaheldin, Elkatatny, 2018. New insights into porosity determination using
trends, challenges, and scenarios for the future. Energy and AI. https://doi.org/ artificial intelligence. Petrolum 4, 408–418.
10.1016/j.egyai.2020.100041. Saputelli, L., Malki, H., Canelon, J., 2002. A critical overview of artificial neural
Duan, Y.X., Wang, Y.F., Sun, Q.F., 2020. Application of selective ensemble learning network applications in the context of continuous oil field optimization. In:
model in lithology-porosity prediction. Sci. Technol. Eng. 20 (03), 1001–1008. Annual Technical Conference and Exhibition, 29 September-2 October. Society
Fan, Z.C., 2016. Evaluation and prediction of fractures in volcanic reservoir of of Petroleum Engineers, San Antonio, Texas, pp. 22–31.
Huangshatuo area. Jilin Geol. 35, 16–20. Sebakhy, E., Emad, A., 2009. Forecasting PVT properties of crude oil systems
Fu, J., Fan, L., Liu, X., et al., 2019. New progresses, prospects and countermeasures of based on support vector machines modeling scheme. J. Petrol. Sci. Eng. 64,
natural gas exploration in the Ordos Basin. China Petrol. Explor. 24, 418–430. 25–34.
Ge, C., Wang, F.Y., Gu, H.O., 2019. Tectonic discrimination based on convolution Shen, Xinzhe, 2018. Oil and gas well test design and real-time optimization
neural network and big data of volcanic rocks. Earth Sci. Front. 26, 22–32. decision-making system Mater Dissertation Research Paper. China University of
Gidh, Y., Purwanto, A., Ibrahim, H., 2012. Artificial neural network drilling Petroleum (Beijing).
parameter optimization system improves ROP by predicting/managing bit Shi, W.Y., Yao, Y.D., Cheng, S.Q., 2018. Transient pressure behavior of acid fracturing
wear. In: Intelligent Energy International, 27-29 March. Society of Petroleum well in tidal flat carbonate reservoirs. In: Kingdom of Saudi Arabia Annual
Engineers, Utrecht, The Netherlands, pp. 2–13. Technical Symposium and Exhibition, 23–26 April. Society of Petroleum
Han, R., Tian, C., Li, S.M., et al., 2017. Selection of model variables for pattern Engineers, Dammam, Saudi Arabia, pp. 35–70.
recognition methods and its application in low resistivity pay identification. Shu, H.L., Wang, L.Z., Yin, K.G., 2020. Geological modeling of shale gas reservoir
Well Logging Technol. 41, 171–175. during the implementation process of geology-engineering integration. China
Huang, F.X., Xia, Z.Y., Gui, H.B., 2016. The Application of BP neural network to DMT Petrol. Explor. 25, 83–95.
hill metamorphic fracture prediction. J. Eng. Geophys. 213, 483–490. Sun, L.D., 2019. Development characteristics and direction of tight oil and gas in
Jia, C., 2017. Breakthrough and significance of unconventional oil and gas to China. Pet. Explor. Dev. 46, 1015–1026.
classical petroleum geology theory. Pet. Explor. Dev. 44 (1), 1–10. Tiago, C.A., De, A., Denis, J.S., 2012. Risk analysis speed-up with surrogate models.
Jiao, F., 2019. Re-recognition of ‘‘unconventional” in unconventional oil and gas. Pet. In: Latin America and Caribbean Petroleum Engineering Conference, 16–18
Explor. Dev. 46 (5), 847–855. April. Society of Petroleum Engineers, Mexico City, Mexico, pp. 22–30.
Jing, C., Dong, X., Cui, W., Dong, Z., Ren, L., Gu, X., He, Y., Pu, C., Liu, X., 2020. Artificial Tong, X.G., Zhang, G.Y., Wang, Z.M., et al., 2018. Distribution and potential of global
neural network–based time-domain interwell tracer testing for ultralow- oil and gas resources. Pet. Explor. Dev. 45 (4), 779–789.
permeability fractured reservoirs. J. Petrol. Sci. Eng. 195, 107558. https://doi. Wan, X.X., Xie, G.L., Ding, Y.G., 2020. Exploration on geology-engineering
org/10.1016/j.petrol.2020.107558. integration of hard-to-recover reserves in Shengli Oilfi eld. China Petrol.
Juniardi, I.R., Ershaghi, I., 1993. Complexities of Using Neural Network in Well Test Explor. 25, 43–50.
Analysis of Faulted Reservoirs. In: Western Regional Meeting, 26-28 May. Wei, C., Cheng, S.Q., et al., 2020. A hybrid analytic solution for a well with a finite-
Society of Petroleum Engineers, Anchorage, Alaska, pp., 20–31 conductivity vertical fracture. J. Petrol. Sci. Eng. https://doi.org/10.1016/j.
Kharghoria, A., Gonzalez, S., et al., 2019. Application of big data techniques in a petrol.2019.106900.
greenfield heavy oil asset in Kuwait – a feasibility study. In: Kuwait Oil & Gas Wen, S.M., Wen, G.H., Li, X.T., et al., 2018. Application and effect of geology-
Show and Conference, 13-16 October. Society of Petroleum Engineers, Mishref, engineering integration in the exploration and development of Baod CBM field.
Kuwait, pp. 20–34. China Petrol. Explor. 23, 69–75.

8
S. Di, S. Cheng, N. Cao et al. Journal of King Saud University – Science 33 (2021) 101542

Wu, J., Liu, B., Zhang, H., He, S., Yang, Q., 2021. Fault detection based on fully Yao, Y.B., 2014. The path analysis of strengthening the cost budget control of oil
convolutional networks (fcn). J. Mar. Sci. Eng. 9 (3), 259. https://doi.org/ enterprises. Modern Econ. 11, 16–17.
10.3390/jmse9030259. Yin, R.W., 2016. Parameter fitting method for reservoir physical properties based on
Wu, Zh.Y., Mo, X.W., Liu, J.H., et al., 2018. Convolutional neural network algorithm particle swarm optimization and support vector machine. J. Xi’an Polytech.
for classification evaluation of fractured reservoirs. Petrol. Geophys. Prospect. Univ. 30, 699–704.
57, 618–626. Yin, X.Y., Chanjuan, L., Baoli, W., et al., 2017. Pre-stack stochastic inversion based on
Xiao, J.L., Li, K.D., Gao, D.W., et al., 2018. Practice and cognition on zipper fracturing hybrid genetic algorithm. J. China Univ. Petrol. (Nat. Sci. Ed.) 41, 65–70.
of horizontal well group in Jiaoshiba block. Fuling China Petrol. Explor. 23, 51– Yuan, X., 2018. Research on crude oil storage and transportation based on
58. optimization algorithm. Advances in materials electronics II: Proceedings of
Xie, J., Zhang, H.M., She, C.Y., 2017. Practice of geology-engineering integration in the 2nd International Conference on Advances in Materials, Machinery,
Changning State Shale Gas Demonstration Area. China Petrol. Expl. 22, Electronics (AMME 2018).
21–28. Zhang, H., Zhu, J., 2013. The innovation and practice of integration mode for
Yang, Z., Tang, Z.X., Chen, X., 2020. ‘‘Exploring oil inside source kitchen”: main types development of KL3-2 oilfield group. China Offshore Oil Gas 25, 67–70.
of tight oil and progress of geology-engineering integration. China Petrol. Zhao, X.Z., Zhao, P.Q., Li D.P., 2018. Research and practice of geology-engineering
Explor. 25, 73–83. integration in.
Yang, Z., Zou, C., 2019. ‘‘Exploring petroleum inside source kitchen”: connotation Zou, C.N., Yang, Z., He, D.B., et al., 2018. Theory, technology and prospect of natural
and prospects of source rock oil and gas. Pet. Explor. Dev. 46 (1), 181–193. gas. Pet. Explor. Dev. 45, 575–587.