Using Semantic Technology to Auto-generate Reports: Case Study of Daily Drilling Reports
- Martin Giese (Univ. of Oslo) | Jens Ingvald Ornas (National Oilwell Varco) | Lars Overå (Posc Caesar Association) | Inge Svensson (Baker Hughes Inc.) | Arild Waaler (University of Oslo)
- Document ID
- Society of Petroleum Engineers
- SPE Intelligent Energy International, 27-29 March, Utrecht, The Netherlands
- Publication Date
- Document Type
- Conference Paper
- 2012. Society of Petroleum Engineers
- 4.1.5 Processing Equipment, 4.1.2 Separation and Treating, 4.3.4 Scale, 1.12.6 Drilling Data Management and Standards, 1.6 Drilling Operations
- 0 in the last 30 days
- 162 since 2007
- Show more detail
- View rights & permissions
|SPE Member Price:||USD 8.50|
|SPE Non-Member Price:||USD 25.00|
The Daily Drilling Report (DDR), a compulsory report to the Norwegian Petroleum Directorate and the Petroleum Safety Authority (PSA) for operators on the Norwegian Continental Shelf, is an example of a task that typically requires integration of data from a variety of independent sources. Current documentation of DDR (EPIM, 2008) content specifies the reporting format in the form of an XML Schema document (XSD), using textual definitions for the explanation of terms.
We report on a case study that demonstrates the use of semantic technologies for the specification, and possibly also generation, of reports in the Oil and Gas sector based on a case study of Daily Drilling Reports. We discuss the underlying vision, present the case study in some detail and evaluate its results. We furthermore discuss the potential of the method for other kind of reports. Finally we conclude with a few recommendations.
The aim of this case study is to investigate the use of cutting edge semantic technologies for the specification, and possibly also generation, of reports in the Oil and Gas sector. This paper concentrates on the vision and main ideas of the approach. In order to give the reader a concrete picture, we have chosen to go to a certain level of technicality. For further technical details, refer to (Overå, 2010).
A report of the type addressed in this study typically contains data aggregated from one or more sources of structured data, and is generated to conform with a predefined format. It is generally accepted that the quality of the reports crucially depends on suitable reporting formats and unambiguous documentation of content. A popular choice is to use an XML-based format, as is done, e.g., for the Daily Drilling Reports (DDR). In the context of DDR, the XML format serves two functions: first, as a transmission format, second, as a documentation format through the associated XML Schema Definition (XSD), along with textual definitions from PCA RDL (PCA, 2008). XML Schema (W3C, 2001) also allows for schema validation for quality control of the data. An overall illustration of the role of the DDR XSD and the PCA RDL in the generation of DDR reports today is given in Figure 1; this is discussed further in the following section.
However, while XML is well suited as a transmission format, it is likely that the level of precision in the documentation can be increased through the use of technologies that implement languages in the so-called "W3C Semantic Web Stack.?? In addition, these technologies may improve the support for automation of report generation and increase the flexibility for reuse of the report data in other contexts. In particular, the following three W3C-recommended languages are of interest:
• The web ontology language OWL (W3C OWL Working Group, 2009), which can be used to formally articulate definitions and relationships between concepts that are today only semi-formally described in the PCA RDL. Using OWL for this purpose is fully consistent with today's practice of using an XML format for data transmission, but it will replace the documentation function of the DDR XSD and extend the function of the PCA RDL.
• The data model RDF (W3C, 2004) that underlies OWL. RDF is based upon the idea of making statements about resources in the form of subject-predicate-object expressions. In RDF terminology, these expressions are known as triples; a database for storing and retrieving RDF data is called a triple store.
• The query language SPARQL (W3C, 2008), which is designed for querying RDF data.
|File Size||356 KB||Number of Pages||11|