The Launch Vehicle Language Data Model Applied to Analysis Tool Connectivity
Jeremy Vander Kam, Peter Gage, Jared Sohn
January 8, 2004
Abstract
Modern multidisciplinary analysis activities include multiple software tools that share large amounts of data. Consistency of this data and how it is shared between tools becomes central to the accuracy and acceptance of the results of such an assessment. A common data model such as the Launch Vehicle Language (LVL) provides a single set of definitions for shared data. Instances of the LVL are unique to the architectures they express. Automated linking of software tools to and from each unique instance requires a dynamic, yet rigorous linking approach. The ability to express generic linkages that scale with each individual assessment subject is possible using the schemas of the LVL and a semantic linking agent. The presented linking approach applies generic linking specifications to each unique assessment to automatically link tools together.
Introduction
A fundamental issue affecting collaborative multidisciplinary environments is automated tool interfacing. Interface complexity increases as the number of software analysis tools in an environment increases. Linkages between inputs and outputs of analysis tools must be consistent in their definitions of units, ground rules and assumptions on a per-variable basis in their definitions of units, ground rules and assumptions. Inconsistently linking outputs from one tool to the inputs of another tool leads to erroneous analysis results. A linking algorithm must ensure that consistent units and definitions exist when inputs and outputs are linked.
State of the art in collaborative multidisciplinary environments is demonstrated by commercial applications such as ModelCenter, Engineous and AML. These frameworks allow tool users to manually specify linkages between tool variables and then transfer data among tools according to these linkages. While this approach is acceptable for largely static data sets, it becomes infeasible when the assessment data changes in quantity and type. Link re-specification takes time and can have a large impact on the efficiency of the analysis team. Each link re-specification provides an opportunity for the introduction of error. A common Aerospace Engineer, NASA, Member, AIAA Technical Director, Eloret, Associate Fellow AIAA Developer, Analytical Mechanics Associates data model combined with a codified linking approach greatly reduces the need for link re-specification.
The Launch Vehicle Language (LVL) is a data model developed to address the aforementioned issues of multi-tool analysis activities. The LVL was initially created to establish a common data vocabulary for launch vehicle data that could be used by analysts to achieve automated linking between tools.
Download The Complete Paper
(A one-time registration will be requested)
