An Improved Methodology for Gridding Fractured Reservoirs for Simulation

A significant amount of research for gridding of complex reservoirs, including models with fractures, has focused on use of unstructured grids. While models with unstructured grids can be extremely flexible, they can also be expensive, both in configuring, computationally, and visual display. Even with this focus on unstructured grids, most reservoir simulation models are still built on structured grids. Current methods for creating reservoir simulation models with structured grids often involve defining a base grid upfront and then "somehow" inserting one or more Features of Interest (FOI's) into the model. Applied to fractured horizontal wells with many stages it can be extremely difficult to accurately align wells and completions within a pre-existing simulation grid. This work describes and demonstrates a methodology to resolve such issues. This approach changes the order of model design and creation steps. This paper describes the process where FOI's are identified, a base grid is designed around the FOI's, then local grid refinements (LGR's) are defined as desired. Applied to a horizontal well with fractures, the well and completion locations are defined before the detailed grid definition is created. This process is illustrated for generalized FOI's, and then applied to fractured horizontal wells. Formulas for creation of models for wells with evenly space homogeneous completions are presented. Numerical testing and analyses are presented that show the impact of the gridding parameters and various design parameters on performance of reservoir simulations.

Solving First-Order Differential Equations of Z-Numbers’ Initial Value Using Radial Basic Function

In this paper, a method was proposed based on RBF for numerical solution of first-order differential equations with initial values that are valued by Z-numbers. The proposed method consists of two parts. The first part has stated the amount of limitation of the fragmentation solution, while the second part has described the assurance of the first part. The limitation section also has two parts. The first part has included the initial condition of the problem, while the second part has included the RBF network. The confidence interval was also considered as a function based on the probability function, which has calculated the confidence level of the first part (limitation). The RBF network or the radial-base grid network has three distinct layers: the input layer that is the set of elementary nodes (sensory units); the second layer is the hidden layers with high dimensions, in which the output layer that has responded to the network response and the activation patterns used in the input layer. The advantage of using RBF is that the use of this technique does not require sufficient information. It only relies on the domain and the boundary. In an example, we have showed that our proposed approach could approximate the problem with acceptable confidence.

342 Economic evaluation of Jersey steers using an aggressive implant strategy

Our objective was to evaluate the profitability of Jersey steers fed in a Non-Hormone Treated Cattle (NHTC) program compared to those receiving an aggressive implant strategy. Jersey steers (n = 30; initial BW 183.3 ± 43.4kg) were allocated to one of two treatments: IMP (Revalor-200 administered at d0-d70-d140-d210-d280-d350) or CON (negative control) and fed for 420 days to assess economic viability for beef production. Live cattle, carcass, and subprimal prices were collected and consolidated from USDA Mandatory Price Reports for the 2018 calendar year. Data were analyzed via independent t-test. Initial cost was $357.02 per animal. Daily feed and yardage costs totaled $2.58 and $2.13 for IMP and CON, respectively. Live cattle value was greater for IMP cattle (P = 0.02; $1,281.20 vs $1,180.70 hd-1) than CON, which resulted from greater live weight. Base grid value was greater for IMP than CON (P = 0.01; $1,209.80 vs $1,076.00) however quality grade discounts were hefty for IMP (-$73.55) due to low marbling and dark-firm-dry outcomes, whereas all CON (+$37.50) carcasses graded Choice and Prime. Premiums for NHTC averaged $121.60 for CON cattle. Although not different (P = 0.19) CON steers netted greater numerical carcass value when marketed on a value-based grid than IMP steers ($1,178.92 vs $1,063.93). Boxed beef values for six steers randomly fabricated per treatment but did not differ (P = 0.87) between IMP and CON ($1,227.19 vs $1,207.92 hd-1). When marketed live, CON returned greater net production value (P = 0.04; -$69.23 vs -$160.60) than IMP, as well as on the grid (P < 0.01; -$71.05 vs -$377.83). However, net production value did not differ when marketed as boxed beef (P = 0.18; -$42.05 vs -$214.56). Marketing Jersey steer beef as NHTC appears to be more economically viable than an aggressive implant strategy, yet neither treatment yielded profit.

Multi-Goal Path Planning for Robotic Agents With Discrete-Step Locomotion

The paper introduces a new, intrinsically discrete, path planning and collision-avoidance problem, with multiple robots and multiple goals. The issue arises in the operation of the novel Swing and Dock (SaD) locomotion for a material handling system. Its agents traverse a base grid by sequences of rotations (swings) around fixed pins. Each agent must visit an array of goal positions in minimal time while avoiding collisions. The corresponding off-line path-planning problem is NP-hard. We model the system by an extended temporal graph and introduce two integer linear programming (ILP) formulations for the minimization of the makespan, with decision variables on the nodes and the edges, respectively. Both optimizations are constrained and favor idling over detours to reduce mechanical wear. The ILP formulations, tailored to the SaD system, are general enough to be applicable for many other single- and multi-agent problems over discretized networks. We have implemented the ILPs with a gurobi solver. Computational results demonstrate and compare the effectiveness of the two formulations.

Initialization of an ice-sheet model for present-day Greenland

We construct initial conditions for an ice flow model of the Greenland ice sheet (GrIS). GrIS has been losing mass at an increasing rate over the past two decades, and a significant proportion of this loss is due to dynamic thinning of narrow outlet glaciers. We solve an inverse problem to estimate poorly known basal and englacial parameters given observed geometry and surface velocities. A weighted cost function, resolved to 4 km in the interior of the ice sheet and 1 km in regions of fast-flowing ice at the margin, is minimized to find two-dimensional fields for a stiffness factor, which is a coefficient of the effective viscosity, and basal traction coefficient. Using these fields, we run the model under present-day climate to damp large-amplitude, short-wavelength fluctuations in the flux divergence. The time-dependent model uses an adaptive mesh with resolution ranging from 8 km of the base grid to 500 m in areas of fast-flowing ice to capture the behaviour of the main outlet glaciers. The ice discharge calculated from the initial conditions for GrIS and individual glaciers compares well with values calculated from observations.

Digital Library Resources in Grid Environment Organization Model

Digital library resources in the grid environment, effective organizational work to prepare the service of the grid system, the user access to information base. Grid resource heterogeneity, distribution, dynamic characteristics, the paper constructed a global / local layered digital library resources organizational model, this mode to avoid the blindness of the resources found to shorten the resources to find time, through constant monitoring of local resource server, so that the search results according to specific rules arranged, greatly improving the utilization of resources to effectively improve the performance of the grid system