Overall Description

The current method of managing drilling signals across data suppliers and data consumers during drilling operations requires perfectly aligned tag names as well as communication protocols. Work Group 1 is tasked with developing a plug and play methodology that does not require human interface to match tag names. Others are working on standardizing communication protocols.

Using a semantic process, each machine/sensor or other element describes itself (what it is and what it can do) and provides this as real time information. [Cayeux, 2019] This allows adding and removing devices from a rig network without affecting existing devices. The description of every device shall be agreed beforehand using techniques based on the output of this committee; namely by describing itself as interrelated “facts” of the form Subject – Verb – Object. This creates a semantic network containing the topological relationship between each device or element that can be used to generate computer code to bridge between various participants devices/elements. However, new device names may be added at any time ...

Note that the devices/elements could be hardware, like a controller or a sensor, or it could be a computation, an estimate, or non-structured data such as a typed report. An application that needs to access information can query the semantic network for the signals that match certain criteria. When the elements are properly described, an application can understand the data, how and where it is obtained and other metadata. Furthermore, this metadata is stored along with the data for use in the future should users need to reexamine the data years later.

This allows drilling automation systems to:

· Discover and select required data, without prior knowledge of the rig signal set-up

o Get access to all the necessary information about the data

o Automatically configure the data transfer

· Get informed when rig signal set-up is modified

o Reconfigure itself

· Provide information about its own data

o Can be used by other systems

There are some basic definitions that help explain the nature of this project.

The SPE-subcommittee “Drilling Data Quality and Uncertainty” has assembled a list of user cases / user pain points and has begun to examine each in detail. The work is currently incomplete and ongoing. This document will be updated periodically as additional items are examined.

Example: abnormal low hook-load detection and characterization

Scenario: A simple quick drilling event detection application focused on abnormal low hook-load would utilize semantic information.

Context:

• The application can detect an abnormal low hook-load with regards to a threshold

• If there is sufficient information, it can categorize the event between a ledge or a pack-off

• The application should be able to distinguish an event from routine variations in hook load due to sheave friction.

Capability:

• It can make use of:

• Block position, bit depth, bottom hole depth

• Several variants of hook-loads

• Top-drive speed and torque

• Flowrate and SPP

• Several sort of mud density in measurements/calculations

• Downhole ECD both through mud pulse and wired pipe

Data producers:

• Drilling control system

• Drilling fluid provider

• Mud logging provider

• Downhole measurements