=Paper=
{{Paper
|id=Vol-3355/geoonto
|storemode=property
|title=Industrial Geological Information Capture with GeoStructure Ontology
|pdfUrl=https://ceur-ws.org/Vol-3355/geoonto.pdf
|volume=Vol-3355
|authors=Yuanwei Qu,Baifan Zhou,Evgeny Kharlamov,Martin Giese
|dblpUrl=https://dblp.org/rec/conf/semiim/QuZKG22
}}
==Industrial Geological Information Capture with GeoStructure Ontology==
https://ceur-ws.org/Vol-3355/geoonto.pdf
Industrial Geological Information Capture
with GeoStructure Ontology
Yuanwei Qu1,* , Baifan Zhou1 , Evgeny Kharlamov2,1 and Martin Giese2,1
1
Department of Informatics, University of Oslo, Norway
2
Bosch Center for AI, Germany
Abstract
Understanding geological structures plays an essential role in a wide range of industries. Various tools
and software have been developed to support the observation and characterisation of the geological
structures and they produce a large volume of heterogeneous data. Thanks to the development of semantic
technologies, plenty of works have investigated industrial geological data. In contrast, geological sketches,
figure interpretations, and illustrations as the initial data collection and analysis steps have received little
attention. Most structure information is still reported as PDF files with forms, tables, and raster images
with interpretations. We attempt to fill this gap by providing an ontology-based industrial geological
information capture method to allow users to input and store structure information formally. Following
users’ sketches, the corresponding knowledge graphs will be generated. This work aims to make the
first-hand geological structure information findable, accessible, interoperable, reusable (FAIR), and to
support qualitative information reasoning, from a user-driven perspective, and to meet the needs of
digitised industry.
Keywords
Geological information capture, Information modelling, Ontology engineering
1. Introduction
Many industries that nowadays serve as foundations of our society greatly rely on a solid
understanding of geological structures, such as petroleum [1], mining [2], sustainable energy [3],
and construction industries [4]. Geologists use sketches to, e.g., assist hydrocarbon exploration
in the petroleum industry [5], to study regional geological structures in the mining industry [6],
and to to evaluate fluid injection, transport, and storage in the sustainable energy industry [7].
Drawing sketches plays an essential role in representing geologists’ understanding and
conceptualisation of geological structures of the domain [8]. The sketching refers to the
process where relevant geological entities from the field will be presented with certain graphical
elements, while irrelevant entities will be abandoned [9], based on geologists’ domain knowledge.
As such a basic but essential knowledge sharing and recording method, sketching has been
widely adopted by geology from the starting point of this domain. It is considered as a primary
and necessary method to assist the geological study [10].
SemIIM’22: 1st International Workshop on Semantic Industrial Information Modelling, 30th May 2022, Hersonissos,
Greece, co-located with 19th Extended Semantic Web Conference (ESWC 2022)
*
Corresponding author.
$ quy@ifi.uio.no (Y. Qu); baifanz@ifi.uio.no (B. Zhou); evgeny.kharlamov@de.bosch.com (E. Kharlamov);
martingi@ifi.uio.no (M. Giese)
© 2022 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).
CEUR
Workshop
Proceedings
http://ceur-ws.org
ISSN 1613-0073
CEUR Workshop Proceedings (CEUR-WS.org)
Though pen and paper are still geologists’ good friends, the need of digitising the geo-
knowledge, allowing geologists to express knowledge in a machine-readable way and improving
data interoperability has increased significantly. In the geology community and industries,
digitisation is drawing significant attention in activities such as ontological modelling [11, 12],
reasoning [13, 14], knowledge graph construction [15], etc.
However, in contrast to these development, sketching is still a few steps behind the digitisation,
despite its important role as an initial data collection step. Many digitised geological methods
work on visualisation and 3D modelling [16, 17, 18]. Some research bridge the animated
interaction and drawing with ontologies [19, 20], but these cannot fulfil the needs of geologists
and the industries.
To this end, we propose an ontology-based sketching system to allow users to document
geological structure information formally so that computers can process them. This system
contributes the encoding of first-hand geological structure information from a user-driven
perspective, making the information findable, accessible, interoperable, reusable (FAIR), and
suitable for inference, and meeting the needs of digitised industry. We present our on-going
research on the ontology-based system in this paper, and it is organised as follows.
• A discussion of lack of studies on sketches (Section 2).
• A early phase work of the structural geology ontology (Section 3)
• A preliminary system architecture (Section 3) that consists of three layers: the user sketch
interface layer, the domain ontology layer, and the intermediate mapping layer.
• A preview of the evaluation methods(Section 4)
• A general conclusion with our future work (Section 5)
2. Related Work
Industrial Impact of Geological Sketches. Geological sketches have long been used in indus-
try. In the traditional petroleum industry, geologists use sketches to express their understanding
of geological structure in the subsurface seismic to assist hydrocarbon exploration [5]. In the
mining industry, one basic approach to prevent mining subsidence and failure is to study re-
gional geological structures, especially preexisting faults [6]. In the sustainable energy industry,
geothermal or carbon capture and storage require a sound examination of geological fracture
systems to evaluate fluid injection, transport, and storage [7, 21], In the construction industry,
especially in railway and tunnel construction, preventing landslide and tunnel water leakage,
and reducing construction failure are also based on the study of geological structures [22].
Work on Semantification of Geological Data. Relying on semantic technologies, much
work has been done to support and enhance industrial geological data interoperability, such
as ontological modelling, reasoning, knowledge graph construction etc. Various ontologies
have been proposed, such as the Ontology of Fractures [11] and the GeoCore Ontology [12].
Qualitative data-based geological multi-scenario reasoning has been proven its usefulness [13].
The work [13] designed an ontology-driven representation of knowledge to support the machine-
readable geological knowledge interoperability. Based on natural language processing and data
mining, a new approach has been developed to construct geological knowledge graphs [15].
Figure 1: An illustration of the system’s three-layered architecture. The intermediate mapping layer
bridges the other two layers and generates the knowledge graphs.
Another work [14] aims at integrating reasoning rules to understand users’ intentions and
problems to improve information retrieval efficiency.
Lack of Study on Sketches. Some drawing tools are designed as digitised canvases to assist
2D/3D modelling, drawing, and visualisation [16, 17, 18], but these visualisation artefacts
themselves contain no geological information, and are not machine-readable. Most information
is still reported as PDF files with forms, tables, and raster images with interpretations. To solve
this problem, ontology-based image annotation tools have been proposed to add meaning to
images [23, 24]. Besides, some ontology-based approaches have been offered to harmonise
the geological fault detection and geological layer horizon in seismic image interpretation and
integrate with geological time [25, 26]. Wang et al. [27] developed a semi-automated method to
utilise text content and elevation model in old geological paper reports to convert raster images
(e.g.geologic illustrations) into vector profiles at low cost. Ontology-based CAD systems like
FieldLog [28] had been implemented to enable geological database flexibility and balance the
needs between users and industries. Ma et al. [20] built an ontology with animated interactive
functions to improve data interoperability. An ontology-based sketching tool was designed
by Forbus et al. [19] to make sketches, in general, more meaningful, though it was mainly
used for education and cognitive science purpose. Inspired by several approaches with similar
technologies or goals have been published, our work is having one step forward to satisfy the
needs of the geology-relevant industries.
3. Our Sketch-Based System for Industrial Geological
Information Capture
We now introduce our system and the methodology.
System Architecture. Our sketch-based system has a three-layered architecture: the user
sketch interface layer, the domain knowledge ontology layer, and the intermediate mapping
Figure 2: Schematic illustration of part of the tailored ontology in a hierarchical view
layer (Figure 1). The user sketch interface layer is a front-end layer that users draw and input
their geological information. The domain knowledge ontology layer provides formal structural
geology knowledge. The intermediate mapping layer takes the sketch input from the user
sketch interface layer and links it with formal knowledge from the domain knowledge ontology
layer, and generates the knowledge graph based on the mapping.
Process of Knowledge Engineering. As a work of developing an ontology-based sketch
system for generating geologic structural information knowledge graphs, constructing a
suitable ontology for the domain knowledge is the first step. The target domain knowledge
is specified in describing the structural geologic scenarios agreed by domain experts. To do
this, we first collect vocabularies that geologists use from structural geology publications. Due
to the large range of the domain, we concentrate our knowledge acquisition work on geologic
structural fault and its relevant concepts as the foundation.
After collecting vocabularies, together with domain experts, a taxonomy has been built to
contain collected vocabularies. Based on domain experts’ reflections, synonyms, conflicts, and
ambiguities were discussed and sorted out during this process. The knowledge is formalised into
a structure that is precise in meaning. In addition to the correctness of the conceptualisation,
the unified terminologies, to a great extent, agreed by domain experts to balance the conciseness
and clarity of our work. To not reinvent the wheel, the work was in alignment with existing
ontologies e.g. Fracture Ontology [11] and GeoCore Ontology [12] to build object properties,
data properties, and classes of the tailored ontology. With a unified top-level ontology as the
backbone, this designed ontology will allow modifications to adapt different sub-domain usage
of the geology. The schematic illustration of part of the tailored ontology is displayed in Figure 2
with four main classes of entities displayed in a hierarchical view.
Our GeoStructure Ontology. Figure 2 displays the sub-classes of the rock, geological object,
geological process, and geological structure and their relationships in the domain of structural
geologic fault. The geological object is displayed in the central and other three entities by
different object properties. The system will generate corresponding knowledge graphs based
on this ontology following users’ sketching processes. A real-world structural geologic fault
system has been used to examine the ontology’s usability, completeness, and expressivity and
the resulting knowledge graph is discussed in Sect. 4 and illustrated in Fig. 3.
Figure 3: A real world structural geology scenario with targeting geological entities(A), an initial
structural geological interpreting sketch of the scenario (B), and the schematic illustration of part of the
generated knowledge graph of the scenario (C)
User Interface Design and Knowledge Mapping The user interface development is in the
stage of wireframe designing. The challenges include how to bridge the graphic sketch part and
the knowledge based part of the system and make the interface easy for industrial users and
geologists. Using a high-level language to make ontology templates is a potential solution for the
first challenge. Solving the second challenge will require a solid user study and interdisciplinary
discussions.
4. Evaluation and Industrial Applications
System Verification. The verification of the ontology is designed in two parts. (1) Competency
Questions. The competency questions are provided by domain experts, and reflect critical
knowledge that needs to be captured in mainly three aspects: fault type, rock layer composition,
and numeric properties. Questions asking for the type of a given fault, the composition of some
layers, and the strike/dip values of the fault, will use to test the ontology. Such questions are
expected to be answered by a knowledge graph adhering to the designed ontology. In addition
to the ontology and knowledge graph verification, future evaluation will use workshops and
interviews as user experience study methods to evaluate the usability of the system, and
what components are required for the system. (2) Examination of Use Cases. In addition to the
competency questions, real scenarios from on-shore field observation and offshore research
work will be used to examine the ontology’s correctness, completeness, and expressivity and
ensure that it can fully describe the scenarios and approved by domain experts?
Examination: Sketch information capture with semi-automated KG generation. In
the current stage, the generated knowledge graph’s clarity is the main focus. To this aim, we
proposed a real scenario in Fig. 3a to see how the a knowledge graph can describe it. Fig. 3a is a
structural geology scenario with faults and host rocks; Fig. 3b is a simple geological sketch of
the scenario in Fig. 3a; Fig. 3c is an illustration of part of the corresponding knowledge graph.
The knowledge graph describes the fault type, the relationship between faults and host rocks,
the composition of the host rocks, and the core numeric values of this scenario concisely.
Applications High quality information and access efficient are essential for digitised industries.
Over time, semantically annotated sketches will form a searchable library, allowing more
efficient access to information than with traditional text- and image-based publications. The
generated information can also support supervised machine learning and qualitative geological
information reasoning in the future to assist users’ decision making. The applications’ evaluation
will be based on industrial needs to run workshops and demos to evaluate the system.
5. Conclusion and Outlook
Conclusion. This paper presents our ongoing research on providing an industrially relevant
information capture method in structural geology. Based on domain experts’ requirements and
existing ontologies, a preliminary ontology has been tailored for the domain to formalise the
knowledge. The capture system generates knowledge graphs in a semi-automated way based
on users’ sketch input. We use a real geological scenario to validate the usage of our knowledge
model and the generated knowledge graph. The generated knowledge graph is retrievable,
reusable, machine-readable, and suitable for inference.
Outlook. Beyond the current work, the front-end system is under development. Our next goal
is to bridge the front-end graphical system and the back-end knowledge base and demonstrate
the work to more users to evaluate the system’s usefulness. The unified ontology shows
great potential in adopting other disciplines of geology, which leaves the possibility for future
extension.
Acknowledgments
This publication is supported by the European Commission H2020 projects Dome 4.0 (Grant
Agreement No. 953163), OntoCommons (Grant Agreement No. 958371), and DataCloud (Grant
Agreement No. 101016835) and the SIRIUS Centre, Norwegian Research Council project number
237898. We gratefully acknowledge the economic support from The Research Council of
Norway and Equinor ASA through Research Council project “308817 - Digital wells for optimal
production and drainage” (DigiWell).
References
[1] A. Torabi, M. U. Johannessen, T. S. S. Ellingsen, Fault core thickness: Insights from
siliciclastic and carbonate rocks, Geofluids 2019 (2019).
[2] L. Donnelly, A review of international cases of fault reactivation during mining subsidence
and fluid abstraction, Quarterly Journal of Engineering Geology and Hydrogeology 42
(2009) 73–94.
[3] S. Loveless, M. Pluymaekers, D. Lagrou, E. De Boever, H. Doornenbal, B. Laenen, Mapping
the geothermal potential of fault zones in the belgium-netherlands border region, Energy
Procedia 59 (2014) 351–358.
[4] R.-q. Huang, Y.-r. Li, K. Qu, K. Wang, Engineering geological assessment for route selection
of railway line in geologically active area: A case study in china, Journal of Mountain
Science 10 (2013) 495–508.
[5] G. Tari, D. Vakhania, G. Tatishvili, V. Mikeladze, K. Gogritchiani, S. Vacharadze, J. Mayer,
C. Sheya, W. Siedl, J. Banon, et al., Stratigraphy, structure and petroleum exploration play
types of the rioni basin, georgia, Geological Society, London, Special Publications 464
(2018) 403–438.
[6] H. Yin, S. Sang, D. Xie, H. Zhao, S. Li, H. Li, X. Zhuang, A numerical simulation technique to
study fault activation characteristics during mining between fault bundles, Environmental
Earth Sciences 78 (2019) 1–11.
[7] S. L. Philipp, A. Gudmundsson, A. R. Oelrich, How structural geology can contribute to
make geothermal projects successful, in: European Geothermal Congress, volume 30,
2007.
[8] K. Forbus, D. Gentner, B. Jee, B. Sageman, D. Uttal, Drawing on experience: Use of
sketching to evaluate knowledge of spatial scientific concepts, in: Proceedings of the
Annual Meeting of the Cognitive Science Society, volume 31, 2009.
[9] Y. Qu, An ontology-based knowledge model for the deep-marine clastic depositional
system, Master’s thesis, University of Oslo, 2020.
[10] B. D. Jee, D. Gentner, D. H. Uttal, B. Sageman, K. Forbus, C. A. Manduca, C. J. Ormand,
T. F. Shipley, B. Tikoff, Drawing on experience: How domain knowledge is reflected in
sketches of scientific structures and processes, Research in Science Education 44 (2014)
859–883.
[11] J. Zhong, A. Aydina, D. L. McGuinness, Ontology of fractures, Journal of Structural
Geology 31 (2009) 251–259.
[12] L. F. Garcia, M. Abel, M. Perrin, R. dos Santos Alvarenga, The geocore ontology: a core
ontology for general use in geology, Computers & Geosciences 135 (2020) 104387.
[13] C. C. Din, L. H. Karlsen, I. Pene, O. Stahl, I. C. Yu, T. Østerlie, Geological multi-scenario
reasoning, NIK: Norsk Informatikkonferanse (2019).
[14] W. Li, L. Wu, Z. Xie, L. Tao, K. Zou, F. Li, J. Miao, Ontology-based question understanding
with the constraint of spatio-temporal geological knowledge, Earth Science Informatics 12
(2019) 599–613.
[15] Y. Zhu, W. Zhou, Y. Xu, J. Liu, Y. Tan, Intelligent learning for knowledge graph towards
geological data, Scientific Programming 2017 (2017).
[16] M. Natali, T. G. Klausen, D. Patel, Sketch-based modelling and visualization of geological
deposition, Computers & Geosciences 67 (2014) 40–48.
[17] E. M. Lidal, M. Natali, D. Patel, H. Hauser, I. Viola, Geological storytelling, Computers &
graphics 37 (2013) 445–459.
[18] C. Jacquemyn, M. E. Pataki, G. J. Hampson, M. D. Jackson, D. Petrovskyy, S. Geiger, C. C.
Marques, J. D. M. Silva, S. Judice, F. Rahman, et al., Sketch-based interface and modelling
of stratigraphy and structure in three dimensions, Journal of the Geological Society 178
(2021).
[19] K. Forbus, J. Usher, A. Lovett, K. Lockwood, J. Wetzel, Cogsketch: Sketch understanding
for cognitive science research and for education, Topics in Cognitive Science 3 (2011)
648–666.
[20] X. Ma, E. J. M. Carranza, C. Wu, F. D. van der Meer, Ontology-aided annotation, visualiza-
tion, and generalization of geological time-scale information from online geological map
services, Computers & geosciences 40 (2012) 107–119.
[21] F. Cappa, J. Rutqvist, Modeling of coupled deformation and permeability evolution during
fault reactivation induced by deep underground injection of co2, international journal of
Greenhouse Gas Control 5 (2010). doi:10.1016/j.ijggc.2010.08.005.
[22] D. Lin, R. Yuan, Y. Shang, W. Bao, K. Wang, Z. Zhang, K. Li, W. He, Deformation and
failure of a tunnel in the restraining bend of a strike–slip fault zone: an example from
hengshan mountain, shanxi province, china, Bulletin of Engineering Geology and the
Environment 76 (2017) 263–274.
[23] F. I. Victoreti, M. Abel, L. F. D. Ros, M. M. Oliveira, Documenting visual quality controls
on the evaluation of petroleum reservoir-rocks through ontology-based image annotation,
in: Theoretical Advances and Applications of Fuzzy Logic and Soft Computing, Springer,
2007, pp. 455–464.
[24] L. A. L. Silva, L. S. Mastella, M. Abel, R. M. Galante, L. F. De Ros, An ontology-based
approach for visual knowledge: Image annotation and interpretation, in: Workshop on
Ontologies and their Applications, in XVII Brazilian Symposium on Artificial Intelligence
(SBIA), Sao Luis, Brazil (Setember-October 2004), 2004.
[25] P. Verney, M. Perrin, M. Thonnat, J. Rainaud, An approach of seismic interpretation
based on cognitive vision, in: 70th EAGE Conference and Exhibition incorporating SPE
EUROPEC 2008, European Association of Geoscientists & Engineers, 2008, pp. cp–40.
[26] L. Mastella, M. Perrin, Y. A. Ameur, M. Abel, J. Rainaud, Formalizing geological knowledge
through ontologies and semantic annotation, in: 70th EAGE Conference and Exhibition
incorporating SPE EUROPEC 2008, European Association of Geoscientists & Engineers,
2008, pp. cp–40.
[27] B. Wang, L. Wu, W. Li, Q. Qiu, Z. Xie, H. Liu, Y. Zhou, A semi-automatic approach for
generating geological profiles by integrating multi-source data, Ore Geology Reviews 134
(2021) 104190.
[28] B. Brodaric, The design of gsc fieldlog: ontology-based software for computer aided
geological field mapping, Computers & Geosciences 30 (2004) 5–20.