Development of the Finite Element Modeling Markup Language

The finite element modeling Markup Language (femML) effort is addressing the problems of data interpretation and exchange for intraand interapplication interoperability in the Finite Element Modeling domain. This is achieved through the development of an extensible markup language (XML) variant for finite element model data that will permit the storage, transmission, and processing of finite element modeling data distributed via the World Wide Web and related infrastructure technologies. The focus of this work was to utilize the XML's power of semantic encapsulation along with the existing and continuously improving associated technology to develop a dialect for exchanging FEM data across various codes with heterogeneous input format syntactic specifications. The main aspects of a finite element definition have been used as archetypes for defining the XML element taxonomy definitions. Namely, the geometry, the material, and the loading aspects of a structural component specification are used to define the first level elements of the associated Document Type Definition (DTD). The element list has been amended with a behavior element specification that represents the solution data to be exchanged or visualized. Various tools have been developed to demonstrate associated concepts along with the ANSYS set of tools. NOMENCLATURE Data Exchange, Interoperability, Data Integration, Finite Element Modeling, Finite Element Analysis, Internet, World Wide Web, Extensible Markup Language. INTRODUCTION General problems The main problems associated with all computationally assisted data exchange, interchange and integration activities can be approached from multiple points of view depending on the needs at hand. However, there is a global point of view that is common to all industries in need of data exchange. In the engineering industries it unfolds as a need for integration of FEM models encoded in multiple data formats from multiple data sources, with current end-user applications and future data exchange systems between applications. However, data interpretation (semantics) varies from data source to data source and therefore introduces semantic correctness uncertainty that destroys robustness of interoperability between applications and data receptacles in general. We have experienced this issue from a very close distance when the time came to implement the Data Driven Design Workbench (DW) used as a virtual wind tunnel environment for design of composite structures and qualification and certification of composite materials systems [1]. This architecture has evolved through time and the whole environment along with the Dissipated Energy Density (DED) methodology have been utilized in various applications including health prediction and sensor optimization of smart structures [2-4]. Inter-modular data flow invited the implementation of some highly structured data format for satisfying the data transfer requirements. One of the most dominant issues on the development of this environment was clearly the semantic biasing and preference from each one of 1 Copyright © 2002 by ASME the modules as well as the lack of a common representation for the exchanged data. It is exactly for cases like this that eXtensible Markup Language (XML) has been used to develop the finite element Modeling Language (femML). In addition to this general need, the proliferation of the specific needs of particular domains of application generate a science push for solving the data structure and heterogeneity of meaning problems within the pertinent vertical industry context. More specifically, digital content with the WWW as a transport medium is available in many forms i.e. multiple commercial applications, manufacturers data-sheets, materials databases, and research and development electronic publications, neutral and custom file formats etc. The need for collaborative dynamic computing through the WWW, strengthens the push for solving the heterogeneity problem by imposing a demand for distributed applications, for problem solving environments, for virtual design and prototyping and for agent-based applications. On the other hand, the multiindustry support and proliferation of XMLware, the JavaDatabase-XML integration technology, and the XML middleware plethora create a technological pull for the utilization of XML-based solutions. Recognition of the problem by the industry The data exchange, interchange and integration problems have been recognized very early by multiple industries of human activity that entails data transfer. Industrial automation technology has improved dramatically over the past decades. CAx systems ("ComputerAided anything", or: CAD, CAM, CAE, CIM, etc.) have provided engineering applications with high-performance solutions. Integration of these technologies is a major issue for industrial competitiveness. From numerical control (NC) in the fifties, through the first design graphics applications and computer controlled production operations in the sixties, Computer Numerical Control (CNC) and Distributed Numerical Control (DNC) in the seventies, and Flexible Manufacturing Systems (FMS) and solid model-based design workstations in the eighties, automation technology has continued to advance and become more sophisticated in order to meet the individual needs of industry. In terms of horizontal integration the CAD industry has responded to geometry data integration and exchange with multiple specific file format specifications. Examples are ACIS, Parasolid, IGES (flavored & standard), STEP, STL, VDAFS, CATIA, CADDS5 etc. [5] However, as industry moves into the 21 century, a new industrial need is becoming the critical problem to solve: the vertical integration of these diverse automation systems (e.g., CAD, CAM, CIM, CAE) [6]. The complex nature of engineering data may hinder the integration of engineering applications. The major “stumbling blocks" that prevent the effective integration of CAx systems are [7]: 1. Current CAx systems have been designed to input and output data rather than information; and 2. Current CAx tools operate on different levels of abstraction of the mechanical product. Therefore, information (data with meaning) modeling is a major issue for CAx systems integration. Moreover, data has to be transferred between applications. An obvious recognition of the importance of XML for information exchange in general, can be evidenced by the plethora of special XML variants developed by and for many non-CAx industries. Attempts to solve the CAx data exchange problem Non XML Efforts The ever present need for data translation to fit the receiving system's data model has been identified as the dominant problem of application integration. To deal with this problem, the International Standards Organization (ISO) launched the STandard for the Exchange of Product model data STEP (ISO 10303-1 1994) [8], aimed at the representation of all information about a product throughout its entire life cycle. STEP allows different applications to exchange information using a standard format. All data models in STEP are normalized (i.e., in conformity with the normal forms, described in section 2.1.2) and written in EXPRESS (ISO 10303-11 1994) [9], an "object-flavored information model specification language" [10] allowing for the specification of complex data models with multiple inheritance. Relative to finite element modeling efforts, the data exchange problem has been traditionally cast under the framework of the product data exchange (PDE) category for most of the historical efforts. Early data exchange specifications focused primarily on geometrical data. Among these were proprietary specifications like Autodesk's DXF, and national standards such as IGES (United States), SET (France), and VDA/FS (Germany). The most significant of these efforts in terms of FEM data representation, is the AP209 ISO/DIS 10303-209 or the STEP 209 Composite and Metallic Structural Analysis and Related Design standard[11]. STEP is a complex standard with huge-sized documents, and was developed as if it was a database itself, adopting the ANSI/SPARC architecture for database systems [12]. Its most significant characteristic is that it allows transfer of conceptual information content in addition to raw data. The standard is comprehensive and is made out of a very extensive but well structured document series [8]. Perhaps its massive specifications and custom and proprietary related tool availability are its two greatest disadvantages. XML Efforts With the advent of XML and especially since its being adopted as a standard specification by the World Wide Web Consortium (W3C) on 1998 [13], many applications became available very quickly. However, only recently we have seen a utilization of XML technology for the engineering applications industries. 2 Copyright © 2002 by ASME It's indicative of the trend that proprietary products to translate STEP documents to an XML form have already been developed to facilitate STEP document transfer. Some companies have decided to integrate XML technology with their existing product lines. Router Solutions Inc. has recently announced [14] their custom CAx integration solution strategy. eXT, is an XML-based 3D MCAD interoperability standard recently introduced by UGS. In short, UGS is wrapping XML around its Parasolid XT format [15,16]. Autodesk Inc.[17] has also recently announced the new XML/Data Extension tool for its AutoCAD 2000i family of products. The XML/Data Extension is part of a broader Autodesk XML initiative to create a common, open standard for delivering design data to the Web, ensuring compatibility between products in different segments of the design industry, and facilitating e-commerce around design specifications. In addition, because of the cross-platform, cross-industry acceptance of XML, the XML/Data Extension will allow developers to create tools that help designers share their design data with other mainstream business functions such as marketing, sales, operations, and customer support. The aecXML project was initiated in August 1999 by Bentley Systems, Incorporated with the desire that it be a unifying force for progress in the development of a project communications framework for architecture/engineering/construction (A/E/C). Bentley has developed an initial specification for aecXML, a framework of XML-based schemas to facilitate communications related to designing, specifying, estimating, sourcing, installing and maintaining construction products and materials over the Internet. Building on the success of aecXML, Bentley and Bluestone have entered a three-year agreement to develop engineering software solutions based on Bluestone's Sapphire/Web Application Server, Bluestone XML SuiteTM Integration Server, and Bluestone's comprehensive standardsbased, e-business solution [18]. In addition to this proprietary efforts there are four public domain efforts very relevant to the engineering data exchange endeavor. These are the Extensible Scientific Interchange Language (XSIL), the FieldML, the X3D (the successor of VRML) that was just released the summer of 2001 and the MatML work in progress for material properties exchange applications. The Extensible Scientific Interchange Language (XSIL) is a flexible, hierarchical, extensible, transport language for scientific data objects. It has been developed at [19]. The entire object may be represented in the file, or there may be metadata in the XSIL file, with a powerful, fault-tolerant linking mechanism to external data. The language is based on XML, and is designed not only for parsing and processing by machines, but also for presentation to humans through web browsers and web-database technology. It comes with a Java object model that is designed to be extensible, so that scientific data and metadata represented in XML is available to a Java code. There is also a powerful Swing-based object browser called Xlook that is also designed to be extensible. The FieldML is an XML-based language for describing time and spatially-varying fields. It is a part of the Physiome Markup languages effort from the university of Auckland in New Zealand [20]. The X3D file format was created to substitute VRML for web based 3D geometries by the WEB3D consortium [21] is basically an XML version of VRML [22] in order to enhance functionality, portability and leverage the Java-XML resources that have been created to support the e-business industry. It can be thought as an XML-interoperable scene graph architecture and encoding standard. Both of these public formats are very useful to the present effort because they represent an extended body of work capable of dealing with the geometry encapsulation, representation and visualization of FEM geometries. The MatML effort [23] is being coordinated by the National Institute of Standards and Technology, and is driven by the MatML Working Group, whose members include several ASM International Fellows, and members from various cross industry organizations. In the 2001 MatML conference [24] a steering committee has been founded to organize the strategic objectives of MatML. The main goal of this effort is the development of the MatML Document Type Definition (DTD), and associated examples and applications, that will facilitate the transfer, exchange and integration of material properties data related to the needs of most CAx industries. CURRENT STATUS OF FEMML Historical note As described earlier, femML has been developed as a necessary outgrowth of our core research efforts, to solve the structured data intensive exchange problem between modules of our custom applications or even between custom stand alone application such as RCfem [25] and existing legacy commercial applications such as ANSYS [26] and ABAQUS [27]. The idea for its creation was naturally generated in the summer of 1999 and has been evolving ever since. FemML’s development went from the conceptualization to the implementation phase when we searched the XML repositories and found nothing relevant to this. Special encouragement for the final push was the lack of responses when we posted inquiries about the possible existence of such an XML variant on October 6, 2000 at various mailing lists like the XANSYS one for ANSYS users. The only relevant XML variants were XSIL, X3D and MatML, all dealing with a partial collection of issues associated with the FEM data exchange, but none of them was directly dealing with the entirety of the main problem. FemML Definition The finite element modeling Markup Language, is an XML variant designed to facilitate the data transfer, exchange, interchange and integration between finite element modeling 3 Copyright © 2002 by ASME applications and their modules. It is work in progress that has accomplished the creation of a DTD, a SCHEMA and certain FEM code specific file generation and parsing tools. It is in a pre-recommendation stage and our focus is to offer it for public discussion, development and distribution. This is not to say that femML cannot eventually evolve to forming the kernel of a set of technologies that will not be solving the data exchange problem, but it will lead to an alternative way of working with FEM data discretizations. A way that would use just three component technologies: • femML as a transport file format, • an ordinary Relational Data Base Management System (RDBMS) for dynamic data management, • and a visualization module. Such a combination of technologies allows composition and factoring of FEM data for the needs of model synthesis and combination as well as the needs for model decomposition and simplification of the design and prototyping industries. FemML Objectives Despite the fact that femML began as a custom effort specific to the data exchange needs within the context of the activities of our group, the objectives employed to motivate the effort of the femML development were very specific and quite general: • Define a standard for the exchange of FEM data (including product shape, associated FEM models, material properties and analysis results) that will allow a person or a computer application to interpret and use the data regardless of its source or target and regardless of the taxonomic order of the FEA model. This effort minimally corresponds to defining: i. A set of XML Tags, ii. Relationships and constraints on these tags, and iii. Document Type Definition (DTD) or/and Schema • Define and develop a set of examples that follow the standard. • Define and develop a set of tools for the utilization of this standard from and to other applications. • Develop examples of using these tools. • Develop a long-term framework for utilization of legacy RDBMS systems, 3D visualization viewer systems, and light-wait asynchronous processes architectures (i.e. agents), for achieving a truly distributed and transparent capability to utilize FEM techniques in highly functional, economical, and ubiquitous manner. By the term “regardless of the taxonomic order” we mean the development of an XML dialect for FEM data exchange that can accommodate all, or most of the FE varieties, i.e. structured, unstructured, blocked, hierarchical, spectral, stochastic etc. femML Document Type Definition (DTD) The current state of affairs has been progressed into the development of a DTD that can definitely cover all of the taxonomic categories of FEM data, except the stochastic ones. The strategy followed for developing femML’s vocabulary of terms, relationships and constrains as well as the DTD that encapsulated them was a special application of the process described by the Unified Modeling Language (UML) [28] activity diagram given in figure 1. Define Vocabulary Terms Define Relationships and Constraints Analyze Human Factors of Vocabulary