Initial prototype for the creation of non-conventional wells. The well trajectory is composed as a piecewise Bézier curve. The user can edit the control points along the well path.
The volumetric lens renders reservoir cells surrounding the user. This spherical lens is defined based on two variables dynamically set by the user: inner and outer radius. This idea reflects two difficulties engineers face while inspecting reservoir models: 1. Quite often engineers need to analyze internal reservoir structures, such as well trajectories and fluid flow lines. In current commercial software, engineers use parallel cuts (a plane oriented with one of the axes) to see the interior of the grid. This approach, however, does not preserve context. On this matter, a commonly used method to address occlusion is Cutaways*: given that an object of interest is in focus, the method automatically computes an appropriate cut surface that eliminates occluding parts while keeping the contextual information in view. In this sense, the volumetric cell lens was inspired by cutaway strategies: internal reservoirs structures are considered to be the important phenomena in focus, and the user selectively discards occluding parts by setting the lens' parameters (inner and outer radius) while maintaining context by rendering reservoir cells located inside the lens. * de Carvalho, F. et al. Interactive Cutaways of Oil Reservoirs. Graphical Models, v. 84, pp. 1-14, 2016. * Lidal, E. M.; Hauser, H.; Viola, I. Design Principles for Cutaway Visualization of Geological Models. In Proceedings of the Spring Conference on Computer Graphics, pp. 47-54, 2012. 2. Current reservoir models employed by the industry may be in the order of millions of cells. Given the large dataset, iso-contouring algorithms are commonly employed to portay only the reservoir surface; yet, as a reservoir is a volumetric entity, some analysis scenarios demand the inspection of its internal cells. The volumetric cell lens, thus, aims at providing a viable visualization that considers a volume of interest while discarding remaining, unnecessary parts of the reservoir in order to lower computational costs.
Cell probing allows users to retrieve local information of reservoir cell units.
System control can be defined as the user interface in which a command is issued whether to request the system to perform a particular function, to change the mode of interaction, or to change the system state. For instance, on the above video, a graphical slider widget allows time filtering. Whenever a time threshold is set by the user, both the 3D graph and the 2D panel are updated accordingly. On this matter, quite often I design 1-DOF graphical menus to be used as system control techniques due to their success and familiarity to users in standard desktops. Further to this, regarding the placement of the graphical menus, a major factor considered is the user's ability to access them. In this sense, I deem the human body to be a strong spatial reference frame; thus, relative locations of the parts of the body such as a person's hand may significantly enhance menu retrieval and usage. The menus may, for instance, enable eyes-off usage, allowing people to perform system control tasks without having to look at the menu.
Basic manipulations such as translation, rotation, and scale are essential for reservoir analysis. While designing these interactions, two major factors were considered: 1. The need to prevent strain injuries that may be caused by repetitive movements since these basic manipulations are intended for extensive use. This is the reason why a certain manipulation starts with an acceleration calculated based on the controller’s velocity; and the motion continues for as long as the user presses the controller’s button; 2. The relevance to leverage the familiarity people may have with metaphors commonly used in 2D interaction spaces. For instance, it is a commonplace on today’s multi-touch touchscreens to use a “pinch” gesture for zooming in and out in a 2D window. Another popular metaphor is the Arcball technique to handle rotating 3D objects using a mouse, proposed by Ken Shoemake*. These prevalent metaphors were, thus, considered and extended to a 3D interaction space for reservoir manipulation. * Shoemake, Ken, "ARCBALL: A user interface for specifying three-dimensional orientation using a mouse," Proceedings of Graphics Interface (GI), Vancouver, British Columbia, Canada, pp. 151-156, 1992.
This was an initial test using leap motion device for performing basic manipulations for reservoir visualization. The 3D UI followed the idea suggested by Remi Bodin: hands fully opened to enable controls - translation, rotation, and scale. The left hand is used to select a control, while the right hand changes the control value.
This collection of short videos showcase prototypes of features for Petroleum Engineering using different spatial interfaces.